Beyond Borders: Satellite Applications For Humanitarian Emergencies

Beyond Borders:
Satellite Applications for

Humanitarian Emergencies
August 2022
AUTHORS
The following authors wrote this report:

Giulia Costella, David Garcia, David Hodgson, Nicki McGoh,
Stella Chelangat Mutai, Catherine Lilian Nakalembe, Rhiannan Price,
and David Taverner.
For further information please contact Nicki McGoh

nicki@cariboudigital.net
ACKNOWLEDGEMENTS
This report was commissioned by the UK’s Humanitarian Innovation

Hub, which is funded by the UK Foreign, Commonwealth and
Development Office (FCDO).
The primary authors were supported by team members from the

wider project consortium:
• Satellite Applications Catapult: Delia Di Filippantonio, Andy

Norris, Chris Reeve, Yolanda Vazquez
• UNITAR: Einar Bjorgo
Caribou Space acknowledges the support offered by staff from the

UK Humanitarian Innovation Hub:
• Ben Ramalingam
• Mark Beagan
• Tonia Thomas
Caribou Space also thanks the project’s Advisory Group (in organisational
alphabetic order) for their valuable contributions to this report.
• Columbia University: Andrew Kruczkiewicz

• European Space Agency: Christoph Aubrecht
• Global Partnership for Data for Sustainable Development Data:
Linet Kwamboka
• Humanitarian OpenStreetMap Team: Rebecca Firth

• UK Foreign, Commonwealth and Development Office:
Fergus McBean
• UK Space Agency: Ray Fielding

• United Nations OCHA Center for Humanitarian Data:
Leonardo Milano
• World Bank: Bernhard Metz

•
RECOMMENDED CITATION
Caribou Space, Beyond Borders: Satellite Applications for Humanitarian Emergencies,

Farnham, Surrey, United Kingdom. Caribou Space, 2022
Published August 2022
www.caribou.space contact@caribou.space
Contents
Executive summary...........................................................................................................3
Background..................................................................................................................... 9
Humanitarian Innovation Hub................................................................................. 10
Satellite applications.................................................................................................11
Humanitarian emergencies.............................................................................................12
What are humanitarian emergencies?.......................................................................... 13
Research framework................................................................................................. 14
Domains of humanitarian emergencies........................................................................ 14
Humanitarian events............................................................................................... 16
Environmental and situational data for humanitarian decision-making...........................19
Ecosystem of satellite applications for humanitarian emergencies................................... 23
Supply-side actors.................................................................................................... 25
Demand-side actors..................................................................................................28
Enabling environment............................................................................................. 31
Satellite applications...................................................................................................... 32
Use cases of greatest need........................................................................................... 33
Use of satellite applications across the event life cycle....................................................... 37
Use cases by domain..................................................................................................40
Satellite applications industry .........................................................................................45
Satellite Earth Observation (EO)............................................................................... 47
Global Navigation Satellite System (GNSS)................................................................. 52
Satellite Communications (SatComms)........................................................................ 54
Landscape of humanitarian satellite applications.............................................................57
Geography.............................................................................................................. 58
User types............................................................................................................... 61
Supplier types......................................................................................................... 61
Domains and events................................................................................................. 63
Business model........................................................................................................ 65
Scale......................................................................................................................66
Data gaps versus availability..................................................................................... 67
Overcoming barriers to uptake of satellite applications in humanitarian emergencies......68
Barriers to uptake.................................................................................................... 69
Priority barriers for development community interventions............................................. 85
Conclusions....................................................................................................................87
Annex 1: Philippines Typhoon Haiyan Case Study......................................................... 90
Annex 2: East Africa Case Study.....................................................................................91
Annex 3: Methodology and limitations...........................................................................93
Annex 4: Glossary.......................................................................................................... 96
Caribou Space
Caribou Space supports organisations to bridge the space and and sustainable
development worlds by working with governments, space agencies, development
agencies, and private sector space companies. Caribou Space provides:
• Official Development Assistance (ODA) fund and programme strategy: Strategic

recommendations for the design and delivery of ODA programmes.
• Fund management: Large-scale ODA funds (£100M+) and seed-stage funds (£4M+).
• Monitoring and Evaluation (M&E): Design of M&E systems, delivery of process and
impact evaluations, and M&E training.
• Research, communications, and knowledge sharing: Conducting research on market

opportunities, user needs, use cases, and impact of space solutions. Sharing knowledge
publicly of what works and doesn’t work—and why—through diverse communications
channels including press and media, publications, social media, conferences, and
workshops.
Beyond Borders: Satellite Applications for Humanitarian Emergencies
• Programme management: Delivery of complex, multi-country, multi-million-pound

programmes in lower-income countries.
• Product strategy: Supporting strategy for the sustainability and commercialisation of

space solutions for lower-income countries.
• Economic evaluation: Quantification of the economic cases and impacts of space technology.
For further information on Caribou Space see

www.caribou.space.

2
Executive
summary
Executive summary
Background
The UK Humanitarian Innovation Hub is a pilot programme, funded by the UK

Foreign, Commonwealth and Development Office (FCDO) and hosted by Elrha, that
seeks to identify the biggest challenges within and opportunities to address global
humanitarian responses.
The objective of this report is to provide a consolidated view on the current use
of satellite applications in humanitarian settings. Satellite applications are digital
services and products that serve several functions for society, the environment, and
the economy. UKHIH selected a consortium of organisations to author this report,
including Caribou Space, Satellite Applications Catapult (SAC), and the United
Nations Institute for Training and Research (UNITAR).
Overview
This report includes the following:
• A categorisation of supply- and demand-side stakeholders of satellite technology for the

humanitarian sector with examples.
• A consolidated overview of examples where satellite technologies have been used

effectively for humanitarian domains.
• A landscaping of over 500 humanitarian satellite applications.
• Identification of key barriers to the adoption of satellite applications and potential

interventions from the development community to overcome them.
• A summary of two case studies from Asia-Pacific and East Africa to provide real-world
insight from specific regions.
Executive summary
4
Humanitarian emergencies
Humanitarian emergencies are “events or series of events that represent a critical threat
to the health, safety, security, or wellbeing of a community or other large group of people.”
They are widely geographically distributed, often covering large areas and multiple
populations. They are diverse and complex, and require coordinated efforts and
resources across a variety of local, national, and international stakeholders.
Humanitarian emergencies share common characteristics related to the domains in

which they occur and the type of events leading to them.
Satellite applications industry
Satellite applications utilise three types of satellite technology:
• Satellite Earth Observation (EO) is the gathering of information about the physical,
chemical, and biological systems of the planet via remote-sensing technologies.
• Global Navigation Satellite Systems (GNSS) are a constellation of satellites providing

positioning, navigation, and timing (PNT) signals from space. GNSS is used to track
people and physical objects at any time, globally. It is also widely used in humanitarian
emergencies for geo-tagging of relevant issues in the humanitarian context, e.g.,
infrastructure, disasters, damages, conflict incidents, response activities, etc.
• Satellite Communications (SatComms) provide voice and data/internet connectivity in

regions that are not covered by terrestrial mobile networks.
Platform and cloud services, together with technologies like machine learning, are
simplifying access to and use of satellite applications. In parallel, new entrants and
developments in SatComms and Satellite Internet of Things (IoT) have the potential to
enable new and more cost-effective connectivity to people and things in the near future.
Satellite applications
The most persistent and extensive humanitarian data gaps include a lack of information
on affected schools, malnutrition, damaged infrastructure like buildings and roads, and
refugees, internally displaced persons, and persons of concern. Satellites are valuable
in addressing data gaps across all of these areas for every humanitarian emergency.
Satellite applications support all stages of the event life cycle, from Hazard and Risk
Assessment to Recovery. Satellite EO can be particularly powerful in supporting
predictive analytics for slow-developing crises, for example, in the context of food
insecurity predicting weak crop yields months before the harvest season, or to anticipate
violence and understand the drivers of conflict.
The improving awareness and acceptance of satellite applications is leading to broader

Executive summary
adoption by humanitarian actors across a wide range of use cases. Such use cases
include population and poverty mapping, infrastructure mapping and exposure, crop
yield and productivity forecasting, hazardous event detection and severity evaluation,
and many other applications across humanitarian domains.
5
Environmental and situational data for humanitarian decision-making
The World Bank’s Data for Better Lives report highlights how the production
and collection of data enable humanitarian and development outcomes. Satellite
applications, as digital technology, provide a unique source of such data about people
and their surrounding environment—with the potential to fill significant data and
information gaps in humanitarian management and decision-making processes.
Satellite applications provide:
• More timely decision-making through real-time and predictive modelling
• Decision-making supported by greater accuracy of data
• Greater confidence in the decision-making process
• Greater accountability across stakeholders
Data from satellite applications has the specific advantages of coverage, objectivity,
repeatability, thematic detail, speed, and affordability.
Ecosystem of satellite applications for humanitarian emergencies
There is a complex network of stakeholders involved in satellite applications for

humanitarian emergencies; the report provides a comprehensive list of actors on this topic.
All stakeholders can be positioned along a spectrum from the supply-side to the demand-
side, or both in a few cases. The supply-side provides the satellite applications, whereas the
demand-side uses those applications to address their humanitarian emergencies.
Supply-side organisations include:
• Private suppliers: satellite operators and resellers, cloud computing providers, platform/
solution providers, and hardware/software suppliers
• Public suppliers: satellite operators, and analytics providers
• Academia
• NGOs
• Media
• Development Agencies
Demand-side organisations include:
• Governments
• First Responders
Executive summary
• Private Sector
• Academia
• NGOs
• Media
6
• Development Agencies
• A ffected Public
• General Public
Landscape of humanitarian satellite applications
This report includes a landscaping of over 500 humanitarian satellite applications which
identified that:
• 42% of identified humanitarian satellite applications targeted government users.
• 62% of identified satellite applications are “customised” for a specific user.
• The highest number (92) of identified applications are for “food, security, nutrition and
famine” events in Africa.
• Many of the data gaps identified by Humanitarian Data Exchange (HDX) in sectors
such as health and education could be addressed by satellite applications.
Overcoming barriers to uptake of satellite applications in humanitarian emergencies
Thirteen barriers to the uptake of satellite applications in humanitarian emergencies

have been identified. Industry, governments, and the development community are
addressing each barrier to varying degrees. Based on the ability of the development
community to help address the issue, a subset of eight barriers are prioritised for
interventions:
• User awareness and resistance

• Inadequate monitoring and evaluation
• Data availability
• Ethics and privacy
• Technical expertise and skills
• Financing for application development and scaling
• Procurement challenges
• Piloting and duplication

Executive summary
7
• Recommended development community interventions are defined for the eight
prioritised barriers.
Barrier Recommended Development Community Interventions
User awareness & • Document existing examples and case studies

resistance • Share information via an open online knowledge base
• Virtual and real-world knowledge events
Inadequate monitoring • Publish impact evaluations via an online knowledge base

& evaluation • Donors ensure impact evaluations are project deliverables
• Training to supply-side organisations on M&E techniques
Data availability • Online open knowledge base of who is doing what and where
• Donors to fund high-resolution EO data for use by humanitarian organisations
Ethics & privacy ocument and share ethics and privacy best practices
• Document
• Upscale community involvement in data generation through community mapping and validation
exercises
Technical expertise & skills • A playbook of principles and case studies that showcase good practice
• Su
Support regional centres of expertise
Financing for application ppropriate patient capital from development agencies and philanthropic organisations through the
• Appropriate
development & scaling creation of new financing mechanisms
cr
Procurement challenges ining for demand-side organisations on satellite application procurement, processes, and pricing
• Training
• Mechanisms to aggregate demand into bulk procurement agreements across multiple demand-side
organisations
or
Piloting & duplication rmation of who is doing what and where via an online knowledge base
• Information
• Enc
Encouraging shared access to data
Conclusions
Satellite applications are being developed by an increasingly complex web of supply-

side stakeholders and are being applied across a range of use cases in the life cycle
of a humanitarian emergency. However, there is a limited body of evidence to offer
humanitarians guidance on where satellite technology generates the best outcomes
for affected populations due to the identified barriers. Now is the time for public and
private sector stakeholders to act in order to increase the use of and impact derived from
satellite applications in humanitarian assistance.
Executive summary
8
Background
Background
Key points
• The UK Humanitarian Innovation Hub is a pilot programme, funded by the
UK Foreign, Commonwealth and Development Office (FCDO) and hosted by
Elrha, that seeks to identify the biggest challenges within and opportunities to
address global humanitarian responses.
• The objective of this report is to provide a consolidated view on the current use
of satellite applications in humanitarian settings.
• UKHIH selected a consortium of organisations to author this report, including

Caribou Space, Satellite Applications Catapult (SAC), and the United Nations
Institute for Training and Research (UNITAR).
• Satellite applications are digital services and products that serve several
functions for society, the environment, and the economy by deploying
three types of satellite technology: Satellite Earth Observation, Satellite
Communications, and Global Navigation Satellite Systems.
Humanitarian Innovation Hub

The UK Humanitarian Innovation Hub (UKHIH) has been established to reduce
the human impact of conflict and disasters by leveraging UK expertise in science,
technology, and innovation. It is funded by the UK Foreign, Commonwealth &
Development Office (FCDO) and hosted by Elrha, a global humanitarian organisation
and the UK’s leading independent supporter of humanitarian innovation and research.
UKHIH is leveraging world-class UK science, technology, and innovation to support

humanitarian action around the world. It is a convening mechanism to promote
collaboration across the UK and provide a coherent, accessible, and impactful
innovation offer from the UK to the international humanitarian community.
Background
By facilitating a network of UK humanitarian innovators focused on new and improved

processes and technologies, the UKHIH aims to make humanitarian action more
effective and efficient, with a particular focus on data and digital. They convene and
coordinate across UK government departments and attract key communities of interest
10 in the UK and globally.
UKHIH selected a consortium of organisations to support this task, including Caribou

Space, Satellite Applications Catapult (SAC), and the United Nations Institute for
Training and Research (UNITAR).
The objective of this report is to provide a consolidated view on the use of satellite
applications in humanitarian settings.
Satellite applications are digital services and products that serve several functions
for society, the environment, and the economy by deploying three types of satellite
technology.
Satellite Earth Observation
Earth Observation (EO) is the gathering of information about the physical, chemical,
and biological systems of the planet via remote-sensing technologies. EO is used to
monitor and assess the status of, and changes within, the earth’s atmosphere, oceans,
land surface, and built environments.
Satellite Communications
Satellite Communications (SatComms) is the use of satellites to provide

communication links between various points on Earth. It is used to provide voice and
internet connectivity in regions that are not covered by terrestrial mobile networks. This
allows communication between people and enables remote monitoring and operation of
machines, sensors, and equipment (Internet of Things (IoT)).
Global Navigation Satellite Systems
Global Navigation Satellite Systems (GNSS) are constellations of satellites providing

signals from space that transmit positioning and timing data to GNSS receivers. GNSS
are used to track people and physical objects at any time, globally.
Background
11
Humanitarian
emergencies
Key points
• Humanitarian emergencies are “events or series of events that represent a
critical threat to the health, safety, security, or wellbeing of a community or
other large group of people.” They are widely geographically distributed, often
covering large areas and multiple populations. They are diverse and complex,
and require coordinated efforts and resources across a variety of local, national,
and international stakeholders.
• Humanitarian emergencies share common characteristics related to the domains

in which they occur and the type of events leading to them.
What are humanitarian emergencies?

A humanitarian emergency is “an event or series of events that represents a critical threat to
the health, safety, security or wellbeing of a community or other large group of people, usually
over a wide area.” 1
They are broad and diverse; whilst no two humanitarian emergencies are the same, they
can share common characteristics. They are based around an event in time that can be
either slow to emerge (e.g., drought) or rapid (e.g., a flood). They pose a threat to large
groups of people. There are typically well-defined phases of an emergency or disaster
management situation that follow a life-cycle process.
Across these phases, information is necessary to support assessment of what is

happening, where, and how. However, this information may be unavailable, incomplete,
or slow to acquire, especially in the immediate aftermath of a humanitarian emergency.
This report is concerned with understanding common information needs during

humanitarian emergencies and identifying critical elements of these needs that could be
addressed using satellite technology.
13
1 Humanitarian Coalition, ‘What is a Humanitarian Emergency?,’ www.humanitariancoalition.ca/what-is-a-humanitarian-emergency

Research framework
The purpose of the research framework is to define the scope of this report and to
highlight key considerations for the use of satellites in humanitarian decision-making
and risk assessment. The research framework draws on existing concepts, models, and
other frameworks to provide a concise knowledge set and theoretical basis. It is used
consistently and built upon in subsequent chapters, providing structure to the analysis
of the research questions.
No obvious, preexisting, “off-the-shelf ” framework for humanitarian events in the

broadest sense has been identified in surveying existing frameworks with project team
input. For example, a United Nations University (UNU) publications search (terms:
“humanitarian” + “framework”) shows 136 results.2 These publications generally include
coverage of events in only one domain or cluster. Brauch, for example, does provide a
number of useful and relevant multi-domain conceptual frameworks for environment
and human security.3 However, the “causes” (e.g., Land, Air, Water, etc.) relevant to
emergencies are too abstract to classify and analyse individual projects in this work. In
considering the structure for this research framework, the primary concerns are to:
• Determine the high-level scope for this report in terms of which existing “domains,”
e.g., food insecurity, and disasters, should be included in scope for humanitarian
emergencies.4
• Provide a non-exhaustive typology of “humanitarian events” that cause humanitarian

emergencies.
• Identify key stakeholders and their needs in the phases of a humanitarian emergency.
Domains of humanitarian emergencies

A domain is a broad grouping of different humanitarian crises with common
characteristics using simple naming and common terminology.
Domains are used to illustrate and define the scope for the report. Domains are simpler
but less specific than the sector and cluster groupings often used by humanitarian
organisations, such as the sphere response sectors and UN Inter-Agency Standing
Committee (IASC) clusters.5 Domains are often closely related and more than one
may apply to an individual humanitarian emergency. This section describes the most
common domains in humanitarian response where satellite technology has the potential
to support. For the purposes of the research, the domains are shown in Figure 1.
Figure 1: Domains of Humanitarian Emergency in Scope

SECURITY FOOD POPULATION HEALTH

DISASTERS
& CONFLICT INSECURITY DISPLACEMENT EMERGENCIES
14
2 United Nations University, ‘Collections,’ http://collections.unu.edu/adv_search.php

3 Hans G. Brauch, ‘Environment and human security: Towards freedom from hazard impacts,’ UNU-EHS InterSecTions 2 (2005),
http://collections.unu.edu/view/UNU:1856
4 This report does not attempt to define new domains.
5 The Sphere Project, The Sphere Handbook 2018, 6 November 2018, https://spherestandards.org/handbook-2018; UN Office for the Coordination
of Humanitarian Affairs Inter-Agency Standing Committee, ‘Reference Module for Cluster Coordination at Country Level, 27 July 2015,
https://interagencystandingcommittee.org/iasc-transformative-agenda/iasc-reference-module-cluster-coordination-country-level-revised-july-2015
Disasters: Natural hazards that can lead to humanitarian situations include floods,
storm surges, earthquakes, landslides, droughts, wildfires, and extreme temperatures.
These events can lead to a humanitarian crisis if they restrict a large group of people
from accessing fundamental needs, such as food, clean water, or safe shelter. Flooding-
related disasters are the most commonly reported and/or recorded.6
Security and conflict: When the protection of the territorial integrity, stability, and
vital interests of states through the use of political, legal, or coercive instruments
at the state or international level fails, violent conflict and insecurity result. Violent
conflicts lead to death and destruction, the crumbling of weak states, local and
international insecurity, a vicious cycle of underdevelopment, instability, and large-scale
humanitarian crises.7
Food insecurity: Food security means that all people, at all times, have physical, social,
and economic access to sufficient, safe, and nutritious food that meets their preferences
and dietary needs for an active and healthy life. Natural hazards, conflict, and supply
chain inefficiencies can disrupt access and contribute to food insecurity. Existing food
insecurity is often exacerbated by conflict that displaces communities that are unable to
carry out day-to-day activities, such as managing crop fields.
Population displacement: At the end of 2020, United Nations High Commissioner

for Refugees (UNHCR) estimated global displacement surpassed 80 million people.8
Refugees are people who have fled war, violence, conflict, or persecution and have
crossed an international border to find safety in another country.9 An estimated 26
million people are refugees, 68% of whom are from Syria, Venezuela, Afghanistan,
South Sudan, and Myanmar. Further, 39% of refugees are hosted in five countries:
Turkey, Colombia, Pakistan, Uganda, and Germany.10 Internally displaced persons
(IDPs) are those forced to flee their homes, but stay within their own country. IDPs
also include people displaced by internal strife and natural disasters. Over 46 million
people are estimated to be IDPs.11
Health emergencies: Public health emergencies include significant outbreaks of

infectious diseases; epidemics, in which a disease affects a large number of people
within a community, population, or region; and pandemics, when an epidemic spreads
over multiple countries or continents (e.g., COVID-19).12
6 For example, 43% of all disasters between 1995 and 2015 in the CRED dataset were flooding-related; Asia and Africa were most at risk. Beatriz
Ravilla Romero, ‘Improving Global Flood Forecasting using Satellite Detected Flood Extent’ (PhD diss., Utrecht University, 2016); Centre for
Research on the Epidemiology of Disasters, The Human Cost of Weather Related Disasters, 1995–2015, 2015, www.unisdr.org/2015/docs/climatechange/
COP21_WeatherDisastersReport_2015_FINAL.pdf
15
7 UN International Peace Academy, The Security-Development Nexus: Conflict, Peace and Development in the 21st Century, May 2004,
www.ipinst.org/wp-content/uploads/publications/security_dev_nexus.pdf
8 UN High Commissioner for Refugees, ‘Refugee Data Finder,’ updated 18 June 2021, www.unhcr.org/refugee-statistics/
9 Same as above.
10 Same as above.
11 UN High Commissioner for Refugees, ‘Refugee Facts: What Is a Refugee?,’ www.unrefugees.org/refugee-facts/what-is-a-refugee/
12 US Centers for Disease Control and Prevention, ‘COVID-19: Identifying the Source of the Outbreak,’ updated 1 July 2020,
www.cdc.gov/coronavirus/2019-ncov/science/about-epidemiology/identifying-source-outbreak.html
Humanitarian events
A humanitarian event is a potential starting point for a humanitarian emergency and
the need for humanitarian action. It is a specific occurrence that initiates the need for
decision-making or action by stakeholders within identified common processes. This
report is concerned with humanitarian events that encompass one or more hazards
that may precipitate a humanitarian crisis and have significant consequences for those
affected. The term “event” is used broadly to describe circumstances wider than hazards,
which can have context-specific meanings.
The research framework includes events and hazards in the identified domains rather
than an exhaustive review. Humanitarian events are rarely discrete and are often
the result of other related events; for example, an IDP crisis may be precipitated by
conflict or a disaster. The exact definition of scope for a specific event and, ultimately,
responsibility for the response to an event is the role of one or more stakeholders. This
may vary by geography and organisational structure, but this report attempts to identify
commonalities across events. It is increasingly common practice to name individual
instances of a realised hazard to aid communications, e.g., the naming of individual
cyclones by meteorological services. Due to the interrelated and combined nature of
events, the term event cluster is used below.
Hazards
Humanitarian events include “hazardous events” as defined by the United Nations

Office for Disaster Risk Reduction (UNDRR).13 They have location, timing, and
measurable risk for associated hazards. The cluster approach and disciplines such as
Disaster Risk Reduction (DRR) and emergency management require preparedness that
relies on hazard identification and understanding.14
Humanitarian events include newly emerged and changing hazards. Hazards are
often described as natural (e.g., fire) or technological (e.g., radiation), with a socio-
natural hazard (e.g., increased flood risk due to environmental degradation) being a
combination of the two. Caribou Space also noted the use of “Natech,” which refers to a
technological hazard caused by a natural one.15 Hazards are global in nature, occurring
with varying severity and frequency in all countries.
Time is an important factor for events, hazards, and their resulting crises and disasters.
They may be permanent, temporary, or episodic.16
• Permanent hazards, for the purposes of risk management, are always present
(e.g., coastal erosion and climate change).
• Temporary hazards can be long-duration (e.g., slow onset) but have a clear start and end
point (e.g., drought).
• Episodic hazards tend to be shorter in duration but recurring (e.g., hurricanes, winter
storms, tsunamis, and river floods).
16
13 UN Office for Disaster Risk Reduction, ‘Terminology: Online Glossary,’ www.undrr.org/terminology

14 UN OCHA IASC, ‘Reference Module for Cluster Coordination at Country Level,’ revised July 2015, https://interagencystandingcommittee.org/
iasc-transformative-agenda/iasc-reference-module-cluster-coordination-country-level-revised-july-2015
15 UN Office for Disaster Risk Reduction, ‘Natech Hazard and Risk Assessment,’ in Words into Action Guidelines: National Disaster Risk Assessment
Hazard Specific Risk Assessment, Part Three: Hazard Specific Risk Assessment, 2017, 90–98, https://www.undrr.org/publication/words-action-guidelines-
national-disaster-risk-assessment
16 Nirupama Agrawal, Natural Disasters and Risk Management in Canada: An Introduction (Springer, 2018)
Episodic and temporary events are the focus of this research, including early warning
systems, rather than permanent hazards that require continuous, ongoing monitoring.
For similar reasons, this report also considers very slow onset events and monitoring
longer-term changes in the natural environment to be out of scope.
Hazards or events that have not exceeded defined risk thresholds may require
monitoring to determine if they will. Some events may be impossible, or at least
difficult, to identify as hazardous in advance with sufficient detail to mitigate, e.g.,
long-term consequences of a political decision.
Existing typologies and records of hazards and disasters, such as the Centre for
Research on the Epidemiology of Disasters (CRED) Emergency Events Database
(EM-DAT), are essential datasets and schemas.17 However, they are typically more
detailed and granular than those found in common usage by humanitarian actors
and stakeholders. The event types used in this report are derived from the existing
typologies within the UNDRR Hazard Clusters.18
These events relate well to the “disasters” domain, but less so to the other four domains.
Therefore, other sources such as The Sphere Handbook have also been used.19 Caribou
Space do not seek to redefine hazard or event types, but rather to highlight those
practically applicable to this research; therefore, some omission and consolidation have
been made. Omissions include cyber hazards, mental health, and food safety, as these
were felt to be outside the direct scope of satellite tools to address.
Risk
For this report risk means “the probability of an outcome having a negative effect on people,
systems or assets.” 20 We also note the importance of valuing potential outcomes from
risks where measured value may be less tangible (e.g., sustainability) or measured and
defined differently by separate stakeholders.21
Major components of risk include “exposure,” to hazardous events, for example;

“vulnerability,” and its related concepts of autonomy, dependency, care, and
exploitation;22 and “adaptive capacity” or “coping capacity” to respond to and recover
from the impacts of humanitarian emergencies.
An important aspect of risk management and risk reduction is understanding risk

through hazards, exposure, vulnerabilities, and adaptive capacity. A key goal in
understanding risk is to quantitatively measure the likelihood of an event happening
and its impacts and consequences.23 Understanding and measuring risk is fundamental
to the use cases within the life-cycle phase ‘Hazard & Risk Assessment’ and all
subsequent stages (see Table 3: List of Use Cases by Event Life-Cycle Phase). Risk is
complex and changes over time through all the event life cycle phases.24 Therefore, the
importance of also addressing multiple hazards and interrelated and dynamic changes
to risk is increasingly recognised.25
17 Centre for Research on the Epidemiology of Disasters, ‘EM-DAT: The International Disaster Database,’ https://emdat.be
18 UNDRR International Science Council, Hazard Definition and Classification Review, 2020, https://www.undrr.org/publication/hazard-definition-
and-classification-review
19 The Sphere Project, The Sphere Handbook 2018, 6 November 2018, https://spherestandards.org/handbook-2018/
17
20 UNDRR, ‘Understanding risk,’ https://www.undrr.org/building-risk-knowledge/understanding-risk
21 Baruch Fischhoff and John David Kadvany, Risk: A Very Short Introduction (Oxford University Press, 2011)
22 Catriona Mackenzie, Wendy Rogers, and Susan Dodds, eds., Vulnerability: New Essays in Ethics and Feminist Philosophy (Oxford University Press, 2013)
23 Jake Ansell and Frank Wharton, eds., Risk: Analysis, Assessment and Management (Wiley, 1992)
24 UNDRR, Global Assessment Report on Disaster Risk Reduction 2019, 2019, www.undrr.org/publication/global-assessment-report-disaster-risk-
reduction-2019
25 Melanie S. Kappes, Margreth Keiler, Kirsten von Elverfeldt, and Thomas Glade, ‘Challenges of Analyzing Multi-Hazard Risk: A Review,’ Natural
Hazards 64 (2012): 1925–58, https://doi.org/10.1007/s11069-012-0294-2
Compound risks and complex emergencies
Increasingly, we face compounding and interrelated environmental, socioeconomic, and

political crises. The current pandemic, for instance, continues to collide with a number
of other threats to human life and livelihoods.26 When multiple risks occur at the same
time, outcomes become significantly worse. As is often the case at the interrelation
between conflict, forced displacement, food security, and disasters. For example, an
area with ongoing or imminent food insecurity hit by a conflict that reduces access
to markets and resources can be particularly devastating. Related to this is the term
“complex emergency,” which can be defined as a humanitarian crisis in a country,
region, or society where there is a total or considerable breakdown of authority resulting
from internal or external conflict.27
This section has focused on introducing a research framework to outline the wider
humanitarian context and define the scope of the research in terms of priority domains
and hazards.
18
26 A. Kruczkiewicz, et al., ‘Opinion: Compound risks and complex emergencies require new approaches to preparedness,’ PNAS 118, no. 19 (2021):
article e2106795118, https://doi.org/10.1073/pnas.2106795118
27 UNHCR, ‘Coordination in Complex Emergencies,’ 1 September 2001, https://www.unhcr.org/en-us/partners/partners/3ba88e7c6/coordination-
complex-emergencies.html
Environmental and
situational data for
humanitarian
decision-making
Environmental and situational data
for humanitarian decision-making
Environmental and situational data for humanitarian decision-making Beyond Borders: Satellite Applications for Humanitarian Emergencies
Key points
• Satellite applications, as digital technology, provide a unique source of data
about people and their surrounding environment—with the potential to
fill significant data and information gaps in humanitarian management and
decision-making processes.
• The World Bank’s Data for Better Lives report highlights how the production
and collection of data enable humanitarian and development outcomes.
• Satellite applications provide:
− More timely decision-making through real-time and predictive modelling
− Decision-making supported by greater accuracy of data
− Greater confidence in the decision-making process
− Greater accountability across stakeholders
• Data from satellite applications has the specific advantages of coverage,

objectivity, repeatability, thematic detail, speed, and affordability.
The World Bank’s Data for Better Lives highlights how the production and collection
of data enables humanitarian and development outcomes. Individuals, civil society, and
academia gain greater accountability and government and international organisations
improve policymaking and service delivery, whilst the private sector gain increased
business opportunities. However, these positive outcomes come with the risks of greater
criminal activity, political surveillance, market concentration, widening inequality, and
discrimination.
20
DATA ANALYTICS
AND PROCESSING
Greater transparency Individuals Greater accountability

Civil society
Academia Criminal activity, dark net
Reuse
Better policy making

More data on individuals Government and service delivery
Data production
International Development
and collection organisations Political surveillance
Reuse
Production process in firms Increased business opportunities

Private sector
Market concentration
Widening inequality
Discrimination
Figure 2: World Bank: How Data Can Support Development: A Theory of Change28
Humanitarian actors continually make complex decisions in the preparation for, and
response to, humanitarian emergencies. Digital technologies support these decision-
making processes by providing large volumes of high-quality data that enables insight.
The term “digital data for development (D4D)” refers to the use of data generated by
mobile, satellite, and digital devices to inform and strengthen nonprofit, humanitarian,
and public-sector decision-making.29
Satellite applications, as digital technology, provide a unique source of data about

people and their surrounding environment and can fill significant data and information
gaps in management and decision-making processes. Often this data may not exist in
any other form (e.g., in remote or conflict areas). In other cases, satellite applications
may offer data that is more frequent, more detailed, and more affordable than ground-
based data sources, such as traditional maps, census data, bespoke surveys, ground
teams, or even innovative methods like the use of drones.30
Satellite applications allow:
• More timely decision-making through real-time and predictive modelling
• Decision-making supported by greater accuracy of data
• Greater confidence in the decision-making process
• Greater accountability across stakeholders
21
28 World Bank, World Development Report 2021 Data for Better Lives, 2021, www.worldbank.org/en/publication/wdr2021
29 igital Impact Alliance, Leveraging Data for Development to Achieve Your Triple Bottom Line, July 2018, http://digitalimpactalliance.org/wp-content/
D
uploads/2018/07/DIAL_D4D_MNO-Paper2_2018-v9.pdf
30 London Economics, Economic evaluation of the International Partnership Programme (IPP): Cost-Effectiveness Analysis, October 2019,
https://londoneconomics.co.uk/blog/publication/economic-evaluation-of-the-international-partnership-programme-ipp-cost-effectiveness-
analysis-october-2019/
Data from satellite applications has the specific advantages of:
• Coverage: Satellites have global coverage that makes it possible to consistently monitor
vast, remote, and even conflict regions across countries and continents.
• Objectivity: Satellite observations derive from the satellite instrument’s measurements,

which have a known and controlled range of error and are thus less susceptible to many
of the biases detected in other measures of the same phenomena.
• Repeatability: Satellite observations are collected along a periodic orbit of the Earth’s
surface, so they are repeatable and comparable over time.
• Continuity: The continuity of satellite data streams allows time to build experience and
refine the systems that use the data.
• Thematic detail: The range of satellite sensors now available allows for application to a
wide range of humanitarian domains.
• Analysis-ready data: Satellite data is organised and processed according to defined

industry standards and provided in forms that allow immediate further analysis.
• Speed: Increasingly, satellite data is available for use soon (days or even hours) after it
is acquired, enabling stakeholders to receive the derived information they need to act
quickly—critical in, for example, disaster scenarios.
• Affordability: Along with the increase in commercial satellites, there is also an increase
in satellites that allow free and open access to data, such as the Copernicus Sentinel
missions.
22
Ecosystem of satellite
applications for
humanitarian
emergencies
Ecosystem of satellite applications
for humanitarian emergencies
Key points
• There is a complex network of stakeholders involved in satellite applications
for humanitarian emergencies; the report provides a comprehensive list of
actors on this topic.
• All stakeholders can be positioned along a spectrum from the supply-side

to the demand-side, or both in a few cases. The supply-side provides the
satellite applications, whereas the demand-side uses those applications to
address their humanitarian emergencies.
• Supply-side organisations include:
− Private suppliers: satellite operators and resellers, cloud computing

providers, platform/solution providers, hardware/software suppliers
− Public suppliers: satellite operators, analytics providers
− Academia
− NGOs
− Media
− Development Agencies
• Demand-side organisations include:
− Governments
− First Responders
− Private Sector
− Academia
− NGOs
− Media
− Development Agencies
− Affected Public
24
“SUPPLY-SIDE” ACTORS “DEMAND-SIDE” ACTORS
PRIVATE SUPPLIER
Satellite Operator and Reseller Academia Government Academia
Cloud Computing Provider

NGOs First Responder NGOs
Platform/Solution Provider
Hardware/Software Supplier Media Development Agency Affected Public
Other
Development Agency Media General Public
PUBLIC SUPPLIER
Private Sector
Satellite Operator
Analytics Provider
Figure 3: Supply-side and Demand-side Actors
All stakeholders related to the use of satellite applications for humanitarian emergencies
can be positioned along a spectrum from the supply-side to the demand-side. The
supply-side provides the satellite applications, whereas the demand-side uses those
applications to address their humanitarian emergencies.
Supply-side actors
The humanitarian ecosystem is composed of an inherently heterogeneous mix of
traditional actors (NGOs and Governments) as well as non-traditional actors (voluntary
technical organisations, e.g., DataKind and the OpenStreetMap community, Private

Suppliers, Academia, Media).
Each stakeholder group has a particular set of interests, motives, assets, skills, and
capacities which directly influence their use of data and value derived from these assets.
Humanitarian actors, their partners, and the Affected Public are producing, capturing,
and accessing growing volumes of data about vulnerable populations and their
environments. These data providers and users together make up a diverse and complex
“humanitarian data ecosystem.”
25
Supply-side actors
Private Suppliers and Public Suppliers represent the majority of providers of satellite
applications for humanitarian emergencies.
Private Suppliers are a diverse group that is further subdivided. Satellite Operators and
Resellers, such as Inmarsat (SatComms) and Maxar (EO), own and operate satellites
and sell data streams from them, either directly or via resellers. Cloud Computing
Providers, such as Google Cloud Platform, Amazon Web Services, and Microsoft
Azure, make it possible to process this data into products by providing on-demand
cloud infrastructure. This software as a service and infrastructure enables all data
analysis and product generation to be implemented in the cloud instead of the user’s
desktop.31 Google Earth Engine (GEE), a cloud-based platform, combines a multi-
petabyte catalogue of satellite imagery and geospatial datasets with planetary-scale

analysis capabilities needed to detect changes, map trends, and quantify differences
on the Earth’s surface. GEE is open access and allows users to conduct analyses that
would otherwise require enormous resources to access, download, store, and analyse.32
Platform/Solution Providers, such as Descartes Labs, build satellite applications that
address specific humanitarian use cases. Examples of Hardware/Software Suppliers
include Garmin GPS for navigation and companies that build large commercial
geosynchronous satellite platforms, e.g., Thales Alenia Space and Airbus Defence and
Space. Other Private Suppliers include miscellaneous private companies in areas such
as agri-business, insurance, mobile operators, and drone operators.
Public Suppliers include two subgroups: first, Satellite Operators, which includes
government space agencies such as the European Space Agency (on behalf of the
European Commission) with their Copernicus constellation and the US National
Aeronautics and Space Administration (NASA). Second, Analytics Providers, such as
mapping agencies, statistical bureaus, land ministries, health ministries, defence and
intelligence agencies, and disaster response agencies use satellite data alongside other
data sources to create and contribute analyses and reports.
Academia and NGOs play roles on the supply-side as data providers and on the
demand-side as users. Academic institutions lead the development of methods required

to extract satellite data to develop products and often make these publicly available for
further use. NGOs rely on insights from this data and play a critical role in collecting
on-the-ground data often needed for training, contextualising, and validating satellite
products.
Media often source and use satellite imagery in news articles to raise awareness among
the public about humanitarian emergencies.
Development Agencies are occasionally on the supply-side, such as the World Bank or
other International Financial Institutions (IFIs), and have internal specialist teams able
to produce satellite applications for humanitarian contexts.
26
31 Catherine Nakalembe et al., ‘A review of satellite-based global agricultural monitoring systems available for Africa,’ Global Food Security 29
( June 2021): article 100543, https://doi.org/10.1016/j.gfs.2021.100543
32 Noel Gorelick et al., ‘Google Earth Engine: Planetary-scale geospatial analysis for everyone,’ Remote Sensing of Environment 202 (December 2017): 18–27,
https://doi.org/10.1016/j.rse.2017.06.031
table 1: Supply-Side Stakeholder Groups
Stakeholder Group Subgroup Examples
Satellite • Airbus, Capella Space, Earth-i, GHGSat Inc., Planet Labs, Inmarsat, Maxar Technologies,
Operators OneWeb, Satellogic, Telespazio
& Resellers
Cloud • Alibaba Cloud, Amazon Web Services, Google Cloud Platform, IBM Cloud, Microsoft
Computing Azure, Oracle, Salesforce, SAP Cloud Platform
Providers
Platform/ • Google Earth Engine, Microsoft Planetary Computer, Orbital Insight, Descartes Labs,
Private Suppliers Solution CrowdAI, Ecometrica, MapBox
Providers
Hardware/ • Hardware: Garmin, Thales Alenia Space, Gilat Satellite Networks, Airbus
Software • Software: Harris (ENVI), ESRI, PCI Geomatica
Suppliers
• Open access: QGIS
Other • Agri-business, Disaster Monitoring Constellation (DMC), generalist consulting firms,

insurance companies, mobile operators, and drone operators
Satellite • European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT),

Operators European Space Agency (ESA), National Aeronautics and Space Administration (NASA),
South African National Space Agency (SANSA), India Space Research Organisation (ISRO)
Analytics • Mapping agencies: Instituto Nacional de Estadística y Geografía (INEGI),

Public Suppliers Providers Geoscience Australia
• Innovation units: Country Directorate of Science, Technology and Innovation (DSTI) in
Sierra Leone, CONCYTEC-Peru’s National Council of Science, Technology, and Innovation
• Statistical offices: UK Office for National Statistics, National Bureau of Statistics
• Defence and intelligence open source projects: xView Challenge, SpaceNet
• Research institutes: Norwegian Centre for Humanitarian Studies, Canadian Research Institute on
Humanitarian Crisis and Aid (OCCAH), African Centres for Disease Control
Academia
• Universities: University of Geneva, Harvard Humanitarian Initiative, Nanyang Technological University
• Think tanks: Chatham House, Freedom House, Center for Strategic International Studies (CSIS)
• Local NGOs: OpenStreetMap (OSM) communities, Map Kibera, iCitizens

• International NGOs: Doctors Without Borders, International Rescue Committee, CARE International,
Mercy Corps, Lutheran World Relief
• Regional organisations: Asia Disaster Preparedness Center (ADPC), Regional Centre for Mapping of
NGOs Resources for Development (RCMRD)
• UN agencies: United Nations Office for Disaster Risk Reduction (UNDRR), United Nations High
Commissioner for Refugees (UNHCR)
• Human rights organisations: Amnesty International, Human Rights Watch
• Initiatives: Digital Earth Africa
• Local journalists: Associated Press

Media
• International journalists: The Guardian, Al Jazeera, Devex, The New Humanitarian
Development Agency • International Financial Institutions: World Bank, Asian Development Bank
27
Demand-side actors
The diverse community of “data providers” is met by an even more diverse community
of global humanitarian data users. Many of these humanitarian actors focus on specific
domains or event types, as discussed in the previous chapter, while others are better
categorised by their roles and workflows during an event.
Government includes senior members of government, such as ministers and

national coordinators, e.g., Pacific Community (SPC) and US Federal Emergency
Management Agency (FEMA). They liaise with other government agencies and
lead teams overseeing analysts and first responders. They engage with the public by
directly briefing the media and public. They also coordinate funding requests from
Development Agency(s). Government also includes mapping agencies, e.g., UK
Ordnance Survey or Survey of India, and defence and intelligence, e.g., Ministries of
Defence. They provide other parts of government with analysis to support decision-
making. They also produce detailed reports for media briefings.
First Responders include emergency services, military response units, and in-country
aid agencies. They provide support in terms of evacuations, search and rescue, and
emergency food, shelter, and medical supplies.
Development Agencies include bilateral and multilateral donors, IFIs, private

foundations, and corporate philanthropies. They provide funding and assistance to
governments, NGOs, and the private sector.
Media include local and international journalists. They report on the humanitarian
emergency to the broader public.
Private Sector includes insurance and reinsurance firms that provide cover against asset
losses, e.g., infrastructure, vehicles, or crops, in the event of a humanitarian emergency,
e.g., a disaster. It also includes multinational companies like Coca-Cola and Nestlé that
have supply chains across the world, which are affected by humanitarian events.
Academia plays roles on both the supply-side as data providers and also as demand-side
users.
NGOs include local and international NGOs and human rights organisations. They
often provide on-the-ground support in the case of humanitarian events, including
providing basic needs such as shelter, water, and healthcare.
Affected Public includes affected communities who play a role in working with
humanitarian stakeholders, especially those on the front lines, to direct and prioritise
assistance.
General Public includes the local and international public. They are government
constituents and influencers, and in some cases contribute to philanthropic funding
appeals to support NGOs in their humanitarian response.
28
table 2: Demand-Side Stakeholder Groups
Humanitarian
Stakeholder Group • Examples
Government • Ministers
• National coordinators: National Disaster Risk Reduction & Management Council (NDRRMC),
Civil Contingencies Secretariat, Pacific Community (SPC)
• Regional: African Risk Capacity
• Mapping agencies: Ordnance Survey, Survey of India, National Geographic Department of Lao (NGD)
• Defence and intelligence: Ministries of Defence, United States Africa Command (AFRICOM)
First Responders • Emergency services: fire, police, medical

• Military response units: Army Corps of Engineers, military medical staff
• In-country aid agencies (local and international): Red Cross, United Nations High Commissioner for
Refugees (UNHCR), United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA),
World Food Program (WFP), Médecins Sans Frontières (MSF)
Development Agencies • Bilaterals and multilaterals: UK Foreign, Commonwealth and Development Office (FCDO),
Japan International Cooperation Agency (JICA)
• IFIs: World Bank, International Finance Corporation (IFC), Asian Development Bank,
International Development Bank (IDB)
• Private foundations: Bill and Melinda Gates Foundation, Rockefeller Foundation
• Corporate philanthropies: Mastercard Foundation
Media • Local journalists: Associated Press

• International journalists: The Guardian, Al Jazeera, Devex, The New Humanitarian
Private Sector • Insurance companies: Swiss Re, Munich Re

• Multinational companies: Coca-Cola, IKEA, Alibaba, Starbucks, Nestlé
Academia • See examples in Table 1
NGOs • See examples in Table 1
Affected Public • Communities directly affected by humanitarian events, who are also volunteers and
crowdsourcing contributors
• Vulnerable groups like refugees and IDPs, slum dwellers, farmers, children, etc.
General Public • Broader national or international public following news of the humanitarian event
• Volunteers and crowdsourcing contributors, e.g., OpenStreetMap volunteers
29
Humanitarian data roles
Beyond thinking about these actors by their traditional stakeholder groupings, e.g.,
Government, Development Agency, etc., it can also be helpful to understand these
groups vis-à-vis their interactions with data. In their paper, Haak et al. present four
different stakeholder data-oriented roles involved in developing, disseminating, and
using a “toolbox” of information derived from public and private sources of data:33
• Initiator/Coordinator, which is the leading agency, most likely a United Nations

agency (e.g., UNHCR, World Food Programme (WFP)), setting up the humanitarian
data ecosystem. In the toolbox context, this party should initiate, coordinate, and
support the development of the toolbox in a country, bringing all relevant actors
together and incentivising them to facilitate data-sharing and use the toolbox.
• Local Lead, which is a national body (e.g., Department of Disasters, Department of

Refugees, Food Security Division) that becomes the local manager after the Initiator/
Coordinator has set out the initial data ecosystem responsible for locally promoting
participation. This role is important to ensure the sustainability of the ecosystem.
• Data Suppliers are responsible for sharing relevant data.
• Toolbox Users are humanitarian or government agencies who should use the toolbox
(i.e., the data processed into understandable information) and incorporate the outputs
into their operations and decision-making.
In some cases, a given stakeholder group can play multiple roles. For example,
governments are often Local Leads, Data Suppliers, and Toolbox Users; their primary
data roles usually vary across departments or agencies. As Local Leads, governments
are on the front lines of a humanitarian crisis and coordinating response efforts. As
Data Suppliers, governments have made significant investments in providing free and
open, continuous and consistent satellite data streams.
In recent years, these governments and the agencies that design and operate satellites
have more actively engaged with the broader humanitarian community to ensure
optimal return on taxpayer investment by encouraging applications to maximise the

societal benefit of the data. For example, the US Geological Survey touts the economic
value of Landsat for the US taxpayer, claiming an estimated US$3.45 billion in
economic benefits to users worldwide in 2017. Similarly, the European Space Agency
showcases the benefits brought by the usage of the Copernicus Sentinel satellite’s data
to society, environment, and economy through a bottom-up assessment of traceable
impacts along selected value chains.34
As Toolbox Users, government agencies leverage public and private satellite data
alongside other sources to inform their analyses, findings, and reports. For example,
national statistical offices use satellite imagery to streamline enumeration campaigns
and more efficiently complete censuses.35
30
33 Elise Haak et al., ‘A framework for strengthening data ecosystems to serve humanitarian purposes,’ Proceedings of the 19th Annual International
Conference on Digital Government Research: Governance in the Data Age (May 2018): article 85, https://doi.org/10.1145/3209281.3209326
34 US Geological Survey, ‘Landsat’s Economic Value to the Nation Continues to Increase,’ 4 March 2021, www.usgs.gov/center-news/landsat-s-
economic-value-nation-continues-increase; European Commission, Copernicus Market Report, February 2019, www.copernicus.eu/sites/default/files/
PwC_Copernicus_Market_Report_2019.pdf
35 Sarchil Qader, ‘Semi-automatic mapping of pre-census enumeration areas and population sampling frames,’ Humanities and Social Sciences
Communications 8 (2021): article 3, https://doi.org/10.1057/s41599-020-00670-0
Enabling environment
Surrounding the supply- and demand-side is an environment that is required to enable
and make best use of satellite applications. Components of the enabling environment
are listed below and detailed later (see section Overcoming barriers to uptake of satellite
applications in humanitarian emergencies):
• Policy and regulation includes satellite-specific policies and regulations as well

as those covering related topics, like data privacy e.g., European Union’s General
Data Protection Regulation (GDPR), international charters (e.g., Disaster Charter),
cybersecurity, data hosting, etc.
• Education and skills encompasses the general awareness, domain expertise, and
technical skills needed to produce or use satellite applications for humanitarian
applications.
• Finance involves the financial organisations providing funding to both supply- and
demand-side organisations, and to other parts of the enabling environment, to facilitate
the development and use of satellite applications.
• Digital/IT infrastructure accounts for the associated technologies and data required to
leverage satellite data streams, including but not limited to cloud computing, internet
connectivity, ground stations, geospatial software, mobile applications and tools, and
other data sources to enhance satellite-derived insights.
Figure 4: Enabling Environments
Dig
tion ita
l In
gula fr
e
r
as
&
tru
icy
ctu
Pol
re
SUPPLY Enabling DEMAND

Environment
Edu
ca
e
nc
ti o
n
& na
sk
ill s Fi
31
Satellite
applications
Key points
• The most persistent and extensive humanitarian data gaps include a lack of
information on affected schools, malnutrition, damaged infrastructure like
buildings and roads, and refugees, internally displaced persons, and persons
of concern.
• Satellites are valuable in addressing data gaps across all of these areas for
every humanitarian emergency.
• Satellite applications support all stages of the event life cycle, from Hazard
and Risk Assessment to Recovery. Satellite EO can be particularly powerful
in supporting predictive analytics for slow-developing crises, for example,
in the context of food insecurity predicting weak crop yields months before
the harvest season, or to anticipate violence and understand the drivers of
conflict.
• The improving awareness and acceptance of satellite applications is leading

to broader adoption by humanitarian actors across a wide range of use cases.
• Use cases include population and poverty mapping, infrastructure mapping

and exposure, crop yield and productivity forecasting, hazardous event
detection and severity evaluation, and many other applications across the five
humanitarian domains.
Use cases of greatest need

At the start of 2021, Humanitarian Data Exchange (HDX) estimated that just 51%
of relevant, complete crisis data is available across 27 categories of humanitarian
operations.36 Population and socio-economy is the most complete data category, while
health and education were the least complete categories. Figure 5 plots HDX’s data
gaps against the most relevant satellite-enabled use case from those listed in this report.
33
36 Sarah Telford, ‘The State of Open Humanitarian Data,’ OCHA Centre for Humanitarian Data, 10 February 2021, https://centre.humdata.org/the-
state-of-open-humanitarian-data-2021/
Figure 5: Completeness of Data Required Across Humanitarian Operations (left) Mapped to Satellite Use Case (right)
Affected School Infrastructure mapping and exposure

7 15 78
Severe Acute Malnutrition Rate Crop yield and productivity forecasting

19 4 77
Affected Areas Hazardous event detection and severity

22 11 67
Global Acute Malnutrition Rate Crop yield and productivity forecasting

30 11 59
Transportation Statues Infrastructure mapping and exposure

26 19 55
Refugees & Persons of Concern Refugee camps

37 11 52
Damaged & Destroyed Buildings Infrastructure mapping and exposure

19 30 51
Baseline Population by Age & Sex Refugee camps

52 4 44
Returnees Refugee camps

48 11 41
Food Security Crop yield and productivity forecasting

70 30
Humanitarian Profile Locations Humanitarian stakeholder coordination

56 15 29
Humanitarian Access Emergency response, & Infrastructure mapping and exposure

11 63 26
Administrative Divisions Population and poverty

74 26
Humanitarian Needs Population and poverty, & Hazardous event detection and severity
74 7 19
Internally Displaced Persons Refugee camps

78 4 18
Baseline Population Population and poverty

67 19 14
Poverty Rate Poverty Rate Population and poverty

89 11
Education Facilities Infrastructure mapping and exposure

11 85 4
Populated Places Population and poverty

37 59 4
Airports Infrastructure mapping and exposure

7 93
Health Facilities Infrastructure mapping and exposure, & Microplanning for health interventions
22 78
Roads Infrastructure mapping and exposure, & Microplanning for health interventions
41 59
3W – Who is doing What Where? Humanitarian stakeholder coordination

74 26
Conflict Events Hazardous event detection and severity, & Evidencing conflict-related violence
96 4
Casualties Hazardous event detection and severity

100
34 Funding None applicable
100
Food Prices Crop yield and productivity forecasting

100
Complete Incomplete No data

While data on affected schools is largely missing (up to 78%), satellite applications help
fill this gap when used for pre-event infrastructure mapping (exposure) and post-event
damage assessment. Before and after imagery helps identify damaged or destroyed
infrastructure, allowing humanitarian actors to assess the near real-time status of
schools and whether they have been affected. This satellite-data-driven approach could
address data gaps related to schools, as well as transport, damaged and destroyed
buildings, and health facilities.
Data on severe acute malnutrition rates is also largely unavailable. However, combining
detection, extent, and severity of events with population and poverty mapping and
forecasts of crop yield can improve these estimates. While satellites cannot directly
observe nutrition levels, they can offer useful proxy information that, when combined
with other data sources, closes some of this data gap. While there has been extensive
usage of satellite imagery for crop models, a focus on nutrition is relatively early-stage
research.37
Across most data categories, there were uniform calls for improving data quality, as well
as access to persistent, real-time data and information streams.38 Beyond types of data
needed, the following requirements emerged from this study, many of which are closely
related:
• Needs assessments and targeting humanitarian operations39
• Democratisation of access to data, especially for affected communities 40

• Monitoring and evaluation of the efficacy of interventions 41
• Predictive analytics42
Needs assessments are critical during humanitarian crises, as they provide baseline
information (e.g., the number of people affected) upon which situation monitoring,
response planning, and severity assessments are based. Needs assessments provide
evidence for the strategic decisions that need to be made, such as where to allocate or
where to deliver aid. In some cases, the availability of such data has been constrained
due to limited access to insecure and remote areas. Moreover, many governments do not
have access to adequate data on their entire populations. This is particularly true for the
poorest and most marginalised communities. Satellites can help provide comprehensive
population data regardless of insecurity or distance from urban centres.
37 Jeremy Irvin, Dillon Laird, and Pranav Rajpurkar, ‘Using Satellite Imagery to Predict Health,’ 2017, http://cs231n.stanford.edu/reports/2017/pdfs/559.pdf;
Elizabeth Bondi et al., ‘Mapping for Public Health: Initial Plan for Using Satellite Imagery for Micronutrient Deficiency,’ KDD 2020 Workshop on
Humanitarian Mapping, 2020, https://teamcore.seas.harvard.edu/files/teamcore/files/kdd_humanitarian_mapping_workshop_2020_4.pdf
38 Henry Lewis and Gary Forster, Humanitarian Research Brief 4: Data Use Capacity in Protracted Humanitarian Crises, Humanitarian Data
Transparency Series Brief 4, Publish What You Fund, June 2020, https://www.publishwhatyoufund.org/wp-content/uploads/dlm_uploads/2020/06/
Humanitarian-Research-Brief-4.pdf
39 Same as above; United Nations, ‘Big Data for Sustainable Development,’ www.un.org/en/global-issues/big-data-for-sustainable-development
40 Kees Boersma and Chiara Fonio, eds., Big Data, Surveillance and Crisis Management (Routledge, 2017); Tina Comes, Kenny Meesters, and Stina
35 Torjesen, ‘Making sense of crises: The implications of information asymmetries for resilience and social justice in disaster-ridden communities,’
Sustainable and Resilient Infrastructure 4, no. 3 (2019): 124–36, https://www.researchgate.net/publication/321691714_Making_sense_of_crises_the_
implications_of_information_asymmetries_for_resilience_and_social_justice_in_disaster-ridden_communities
41 Comes, Meesters, and Torjesen, ‘Making sense of crises.’
42 UN Committee of Experts on Global Geospatial Information Management, Future Trends in Geospatial Information Management: The Five to Ten
Year Vision, August 2020, https://ggim.un.org/documents/DRAFT_Future_Trends_report_3rd_edition.pdf; Kevin Hernandez and Tony Roberts,
Predictive Analytics in Humanitarian Action: A Preliminary Mapping and Analysis, K4D Emerging Issues Report 33, Institute of Development
Studies, June 2020, https://opendocs.ids.ac.uk/opendocs/bitstream/handle/20.500.12413/15455/EIR33_Humanitarian_Predictive_Analytics.
pdf ?sequence=1&isAllowed=y
Democratisation of access to data, especially for affected communities, ensures
that humanitarians can better collect data, share information, and validate findings
with communities on the ground. This can inform intervention priorities and
locations to target resources and services. In a context of increasing social and digital
inequalities, technologies run the risk of continuing to benefit more resilient and less
at-risk populations. However, with intentional humanitarian data stewards—people
responsible for championing improved data usage—satellites can help address digital
divides during events rather than exacerbate them. For example, satellite imagery
provides context and maps of every community regardless of geographic spread, size,
wealth, political affiliations, or administrative boundaries. By enabling feedback loops
with affected communities, humanitarians would not only ensure the right to access the
data collected about these populations but also gain the best possible “groundtruth,” or
confirmation of their interpretation of remotely collected data.
Monitoring and evaluation of the efficacy of interventions is another key use

for satellite applications. Government and UN coordinators need more oversight
information, while NGOs and other demand-side organisations require specific
information to design and implement their programmes. Given their revisit rates and
ongoing data streams, satellites are an important source for monitoring and evaluating
humanitarian programmes. During a conflict or disaster, for example, satellite imagery
can evidence which buildings and roads have been damaged, as well as which have been
subsequently rebuilt. Combinations of satellite data sources like EO constellations and
GNSS alongside ancillary data, such as surveys, offer even more complete monitoring
and evaluation.
Predictive analytics, in addition to observed data, allow anticipatory action. By

monitoring sources, patterns, and trends, potential crises can be detected and averted.
Insights from predictive models can improve humanitarian supply chains, reduce
deployment costs, forecast food shortages, suggest migration routes, identify disease
transmission patterns, and track malnutrition trends. The ever-increasing scale of
data collected provides important historical and current observations to support more
accurate forecasts and predictions. Importantly, predictive analytics allows the use of
satellite applications not only for responding to emergencies but also for preventing
emergencies. In this regard, EO can be particularly powerful for slow-developing
crises, for example, in the context of food insecurity predicting weak crop yields months
before the harvest season. Every dollar invested in the prevention of humanitarian
emergencies might save several dollars/pounds as compared to an emergency response.
36
Use of satellite applications across the event life cycle
Satellite applications have a role across all stages of a humanitarian emergency, which
are visualised in Figure 6.
1 Hazard and Risk Assessment – The first stage whereby hazards and risk factors
that have the potential to lead to a humanitarian situation are identified. Hazard
identification involves characterising hazards according to the probability of the hazard
becoming reality and its potential severity.
2 Mitigation – This stage includes any action or sustained effort undertaken to reduce a
hazard through the reduction of the likelihood and/or the consequence component of
that hazard’s risk.
3 Preparedness – This stage includes knowing what to do in the aftermath of an event

and being equipped with the right tools and information to act effectively.
4 Response – The following stage involves actions aimed at limiting injuries, loss of life,
and damage to property and the environment taken before, during, and immediately
after an event. Response processes begin as soon as it becomes apparent that an event is
imminent and last until the emergency is declared to be over.43
5 Recovery – The stage when communities, systems, and lives are rebuilt, repaired, and
otherwise returned to functional conditions.
Table 3 provides use case examples categorised by the humanitarian event life cycle.
These examples are not exhaustive but are indicative of the wide-ranging and persistent
applicability of satellite data across the humanitarian spectrum.
Each use case is categorised as to whether satellite data provides direct value in
providing insights or indirect value because it needs to be combined with other data,
improved with ground validation, and/or integrated into a more sophisticated analysis
or model. For example, predicting and analysing crop yields remotely is a powerful use
case of EO in the humanitarian context, for technical reasons and given the potential
impact on saving lives. In turn, poverty mapping is of course important for the creation
of baselines, but EO can only effectively help in this case if combined with other
datasets and (often complex) analyses, and faces caveats and margins of error.
Figure 6: Multi-stage Life Cycle of Humanitarian Emergencies
Haza
ery rd a
cov nd
Re Ri
s
kA
sse
ssm
ent
onse
Resp
n
a ti o
i ti g
M
37
P re
p ar e d n e ss
43 Damon P. Coppola, Introduction to International Disaster Management, 3rd ed. (Elsevier Science, 2007)
table 3: List of Use Cases by Event Life Cycle Phase
Life Cycle Satellite Data

Phase Use Case Value Domain-Specific Examples
Monitoring nuclear • Monitoring of suspected proliferation installations for the detection

Direct value
facilities of undeclared nuclear facilities and processes
Monitoring prison • Estimating populations at prison camps

Direct value
camps • Evidencing the existence of prison camps in inaccessible areas
Infrastructure • Mapping roads to determine access to resources/services

mapping and Direct value • Reducing seismic risk for buildings
exposure • Quantifying financial exposure of assets
Hazard
& Risk Natural resource
Indirect value • Identifying resource scarcity and likely drivers of conflict
Assessment mapping
Population and • Disease outbreak modelling and hotspot identification

Indirect value
poverty mapping • Identifying coastal communities at risk for flooding and surge
Subsidence and • Warning about potential infrastructure instability

Indirect value
erosion mapping • Prioritising and reinforcing engineering projects to address hazards
Disaster risk financing • Property underwriting and recommendations

Indirect value
and insurance • Predicting how many claims insurers will receive
Crop/livestock • Analysing climate impact on crops and fields’ productivity

Direct value
insurance • Evidencing number of large livestock owned or lost
Rangelands • Forecasting food insecurity in pastoralist communities using livestock loss

Direct value
management
• Evidencing conflict-related cattle rustling
Mitigation Access to finance and

Indirect value • Evidencing good agricultural practices to determine creditworthiness
credit
Early warning • Identifying conditions for disease outbreak, famine, or disasters

Indirect value
systems • Dam monitoring as part of flood early warning systems
Extraction and • Advocating traceability across supply chains

trafficking of illegal Indirect value
resources or humans • Tracking illicit mines, logging, and other trafficked goods
Crop yield and • Early warning systems for crop failure or seasonal decline
productivity Direct value
forecasting • Yield gaps based on potential and observed crop growth
Election polling site • Preventing electoral violence with monitoring of election sites
Direct value
monitoring • Ensuring vulnerable populations have access to polling sites
Preparedness
Micro-planning for • Guiding immunisation campaigns and distribution of vaccines
Indirect value
health interventions • Navigating community health workers to reach every household
• Tracking conflict-sourced goods and disrupting trafficking

Humanitarian supply
Indirect value • Improving humanitarian operations and resource distribution
chain management
• Asset tracking of high-value resources (vaccines, food, water, etc.)
38
Agricultural pest and • Monitoring conditions on-farm and assessing risk for pests and disease
Direct value
disease response • Allocating limited supplies (e.g., pesticides, sprayers) to affected areas
Connectivity for • Providing communications to refugee camps

public or remote Direct value • Coordinating delivery of aid
populations • Communicating in remote functions like border or forest patrol
• Identifying destroyed infrastructure in conflict settings

Damage assessment Direct value • Identifying burned buildings from wildfires
• Evidencing impact of event on natural resources
• Conducting preliminary needs assessment to add to local knowledge

Emergency response Direct value
• Identifying evacuation routes and ingress/egress routes
• Providing cash transfers in emergency settings

Financial payments Direct value
• Connectivity for rural/remote banking
Search and rescue • Identifying indications of human activity or presence along with
Direct value
missions terrain models to focus search areas
• Determining likely source and spread of contamination as well as

Contaminated sites Direct value
at-risk populations
Access routes and • Identifying traversable routes based on debris and roadblocks
Direct value
evacuation plans • Assessing access to markets and trade hubs in rainy season
Response
Public crisis • Evidencing humanitarian need for fundraising appeals
Direct value
awareness • Using imagery in news stories to provide context
• Mapping patients’ movements to limit transmission

Contact tracing Direct value

• Identifying the origins of an outbreak to inform treatment
Humanitarian
stakeholder Direct value • Providing situational awareness to enable stakeholders to coordinate
coordination
• Providing before and after images of destroyed communities and

Evidencing conflict- property destruction
Indirect value
related violence
• Evidencing mass burial sites
Hazardous event • Detecting conflict-related events, natural disasters, crop failures

detection and Indirect value
severity • Forecasting disaster intensity and impacts
• Mapping digital terrain and surfaces to chart flight paths

Guiding drones Indirect value
• Identifying communities with least access for drone deliveries
• Assessing quality of life in camps

Refugee camps Indirect value
• Monitoring camp growth and effects on host communities
Disease prevalence
Indirect value • Predicting the location and spread of diseases
and transmission
Infrastructure • Evidencing progress of reconstructed infrastructure

Direct value
reconstruction • Increasing infrastructure resilience based on reconstruction
Compliance with • Verifying disarmament per peace treaty terms

Indirect value
peace treaties
• Preventing escalation of conflict by proving claims of disarmament

Recovery
Monitoring and • Evidencing baseline conditions
Indirect value
evaluation • Monitoring impacts (e.g., infrastructure rebuilt, camp populations, etc.)
Analysing refugee
resettlement Indirect value • Identifying and planning locations to resettle refugees
suitability
39
Use cases by domain
Satellites provide critical data across all domains of humanitarian activity.
This section provides a broad overview of use cases by humanitarian domain.
Disasters
Spatially disaggregated risk maps showing exposure to hazards and vulnerability to

events and their impacts are derived from satellite data. During hazard identification,
data sources like satellite imagery are critical and are often the only option for mapping
the extent of previous flooding events, as one example, and for modelling flooding
threat thresholds. Moreover, during the response phase, satellite-derived maps provide
information on extent, severity, and impacts post-disaster. The data can inform
appropriate mitigation measures and response plans, as well as optimise evacuation
routes and guide response teams. The data is useful to identify affected communities
and prioritise most-affected areas based on damage assessments, population,
vulnerability, and other variables that are derived from satellites.
Beyond situational information, SatComms systems are often essential to enable

stakeholder coordination in remote regions without mobile coverage or where mobile
networks have been damaged by the event. SatComms systems ensure connectivity for
first responders as well as affected communities.
Early warning systems

Hazardous event detection and severity
2.5%
Emergency response
2.
5 Damage assessment
2.8 % 27.2%
% Subsidence and erosion
3.3
% Infrastructure mapping and exposure
Humanitarian stakeholder coordination
4.2%
Disaster risk financing and insurance
4.7% Natural resource mapping
Population and poverty
5.0% Contaminated sites
13.6% Infrastructure reconstruction — 1.9%
6.1% Crop/livestock insurance — 1.4%
Connectivity for public or — 1.4%
8.1% remote populations
12.2%
Extraction and trafficking of — 1.1%
illegal resources or humans
Other
Figure 7: Frequency of Use Case in Disasters Domain (n =>500)

Security and conflict
Satellites are valuable assets in conflict situations, where access to the affected area
is often problematic for ground teams aiming to conduct conventional field mapping
40
during early phases. Without a physical presence on the ground, satellites can provide
timely intelligence on conflict-related destruction, displaced populations, refugee
and IDP camp growth, and human rights abuses, as well as inform post-conflict
reconstruction and monitoring of peace treaties.
Evidencing conflict-related violence
3.5%
Extraction and trafficking of illegal
15.4% resources or humans
2.
2.
1
1%
Damage assessment
2.8 %
% Monitoring prison camps
2.8
% Public crisis awareness
Infrastructure mapping and exposure
4.9% 14.0%
Infrastructure reconstruction
Nuclear facilities
6.3%
Natural resource mapping
Election polling site monitoring
6.3% Compliance with peace treaties
13.3%
Refugee camps
6.3%
Search and rescue missions
7.0% 7.7%
Population and poverty — 1.4%
Monitoring and evaluation — 1.4%
Other
Figure 8: Frequency of Use Case in Security and Conflict Domain (n =>500)
In some cases, satellites can also help prevent conflict by evidencing de-escalation of
militarisation and adherence to international humanitarian law. As documentation
tools, satellites have the potential to collect important real-time evidence for alleged
war crimes and crimes against humanity. Because these events often occur in
inaccessible environments, over large geographic areas, and/or across long and multiple
time frames, high-resolution satellite imagery, in particular, has value for international
tribunals investigating these crimes.
Food insecurity
Information on crop conditions and yield estimates required to stabilise markets,

mitigate food supply crises, and mobilise humanitarian assistance efforts to evaluate
and target production-enhancing interventions can be derived from EO satellites. For
food insecurity use cases, the consistent revisit rate of satellite constellations allows
for crop monitoring to support early warning systems. Beyond revisit rates, spectral
information in satellite imagery can unlock important dimensions like crop health,
crop-type mapping, soil moisture, suitability maps, and many other food security–
related indicators that support more complex crop performance models, pest and
disease models, food demand, and more. Combining these agricultural datasets with
population and poverty data can be critical for assessing the scale of humanitarian
needs and planning rapid response measures.
In the mitigation phase, satellite data, including both EO and SatComms, can inform
the implementation of disaster risk financing, agricultural insurance, and access to
credit, supporting more resilience in communities especially vulnerable to climate
shocks and stressors.
41
6.0% Crop yield and productivity forecasting
7.2% 34.1% Early warning systems

Agricultural pest and disease response
Crop/livestock insurance
9.6% Rangelands management
Access to finance and credit
Damage assessment — 1.8%
Infrastructure mapping and exposure — 1.8%
10.2%
Hazardous event detection and severity — 1.2%
13.8% Emergency response — 0.6%
12.0% Microplanning for health interventions — 0.6%
Public crisis awareness — 0.6%
Subsidence and erosion — 0.6%
Figure 9: Frequency of Use Case in Food Insecurity Domain (n =>500)
Population displacement
Owing to its continuous spatial coverage and frequency of collection, satellite data, like
optical and synthetic aperture radar (SAR), provides insightful, near-real-time context
for analyses.44 By linking this data with GNSS data, population displacement can be
mapped and monitored.
Data derived from optical imagery, like temporary settlements, cattle, boats, vegetation,
water, and other features, can be used to estimate the number of people displaced or
the extent and status of infrastructure and natural resources in resettlement areas. It
also provides context-specific natural hazard risk assessment for informing mitigation
measures to protect already vulnerable communities.45
Satellite imagery supplements conventional data—in “hard to reach areas” satellite

imagery can be the only source available—to provide detailed insights on local
population patterns without raising data privacy concerns or endangering staff with
unknown risks. Immediate needs are quickly evidenced, and humanitarians can
effectively respond and target scarce resources. Whether monitoring the growth of
refugee and IDP camps or the nearby resources and host communities, satellites can
reduce the need for ground-based surveys.
Combining data on nearby communities, the surrounding environment, livelihoods,

infrastructure, and topography helps target interventions and avert cascading
humanitarian emergencies. Historical satellite data on migratory movements, alongside
various biophysical and socioeconomic indices, can help build machine learning models to
improve forecasting of population displacement and simulate optimal response scenarios.
For governments, satellite imagery is an important tool to monitor border crossings,

assess the origin and cause of displacement, understand the scale of migration, and
prepare to host and eventually resettle refugees.
42
44 International Organization for Migration, ‘Migration Data Portal,’ updated 26 May 2021, https://migrationdataportal.org/data-innovation/
interactive-tool-map-world-internally-displaced-satellite-imagery
45 Nistara Randhawa et al., ‘Fine scale infectious disease modeling using satellite-derived data,’ Scientific Reports 11 (2021): article 6946,
https://doi.org/10.1038/s41598-021-86124-2; Shifeng Wang, Emily So, and Pete Smith, ‘Detecting tents to estimate the displaced populations of
post-disaster relief using high resolution satellite imagery,’ International Journal of Applied Earth Observation and Geoinformation 36 (April 2015):
87–93, https://doi.org/10.1016/j.jag.2014.11.013; A. Heslin, L. Thalheimer, and M. J. Purma, ‘Using Remote Sensing to Measure Displacement and
Assess Environmental Conditions at Host Locations’ abstract, American Geophysical Union Fall Meeting 2019, https://ui.adsabs.harvard.edu/
abs/2019AGUFMGC13G1225H/abstract
Refugee camps
Public crisis awarenesss
Damage assessment
3.1 Evidencing conflict-related violence
% 35.9%
3.1% Emergency response — 1.6%
Natural resource mapping — 1.6%
4.7% Microplanning for health interventions — 1.6%
Extraction and trafficking of illegal — 1.6%
resources or humans
Connectivity for public or — 1.6%
14.1% remote populations
Contaminated sites — 1.6%
Monitoring and evaluation — 1.6%
23.4% Election polling site monitoring — 1.6%
Compliance with peace treaties — 1.6%
Refugee resettlement suitability analysis — 1.6%
Figure 10: Frequency of Use Case in Population Displacement Domain (n =>500)
Health emergencies
Across health emergencies, satellite data serves a number of important functions.

GNSS and GIS are used for the study and forecasting of communicable and
noncommunicable diseases; SatComms for telemedicine and tele-education; satellite
imagery for microplanning interventions; and GNSS to deliver access to healthcare and
safely and efficiently transport patients and medical supplies.
Satellite applications can produce more detailed maps of population demographics

and locations connected by human movements. Data on health facilities, roads, and
populations can be used for real-time tracking of disease spread, planning and control
of contact tracing, and immunisation programmes during epidemics and pandemics.
Spatial models for disease transmission and risk profiles are possible when combining
satellite imagery with GNSS, population density, and clinic data, for example,
identifying areas of malaria risk by identifying breeding grounds via stagnant water
mapping and consequently targeted interventions.
It is also possible to combine these data sources to forecast capacity and demand by
health clinics. Satellite applications can also be used to continuously assess national
health systems and generate health facility readiness scores for COVID-19 response.
While not specific to health, population and poverty mapping are critical to any health
surveillance approach, especially in an emergency setting where a lack of understanding
of health catchment areas can cost precious time and resources. High-resolution
population estimates support the measurement of programme targets, coverage, and
improved programme planning.
Satellite data is valuable for mapping digital elevation models to chart navigation and
guide drones for vaccine delivery to remote clinics; however, this is a newer application,
43 as most cargo drones are still prototypes.46
46 Swiss Foundation for Mine Action, Drones in Humanitarian Action: A guide to the use of airborne systems in humanitarian crises, December 2016,
https://reliefweb.int/report/world/drones-humanitarian-action-guide-use-airborne-systems-humanitarian-crises
Microplanning for health interventions
2.7 %
Disease prevalence and transmission
2.7
2.
%
7% Population and poverty
4.1% Early warning systems

31.1%
4.1%
Contact tracing
Emergency response
5.4%
Public crisis awareness
5.4%
Connectivity for public or
remote populations
5.4% Access to finance and credit
20.3% Hazardous event detection and severity — 1.4%
10.8%
Humanitarian supply chain management — 1.4%
Access routes and evacuation plans — 1.4%
Guiding drones — 1.4%
Figure 11: Frequency of Use Case in Health Emergencies Domain (n =>500)
Cross-cutting use cases

As highlighted by population and poverty mapping, many use cases are valuable
across multiple domains and humanitarian events. For example, mapping damaged
infrastructure due to disaster or conflict can also inform better estimates of the
impacted and displaced populations and where food aid and other resources are
required. Other cross-cutting use cases include connectivity for public or remote
populations and early warning systems.
Satellite applications that support cross-cutting use cases are particularly relevant
for compound risks and complex emergencies. For example, in Lake Chad Basin
environmental degradation, a very high fertility rate, forced displacement, and repeated
herder–farmer conflicts create significant compound risk.47 EO is a powerful tool to
understand the dynamics and interrelation between such factors.
Satellites provide wide-reaching value in answering foundational humanitarian

questions: Who is vulnerable? What help is needed? How can communities in need
be reached? What are likely risks and impacts? Satellite applications offer a consistent,
global common operating picture for stakeholders to consider risks and impacts and
then coordinate to make targeted, effective decisions. Few other sources provide
such valuable information across the entire event life cycle, across so many different
domains, and over an entire global population.
44
47 Personal communication to Caribou Space, 2021.

industry
Satellite applications industry
Key points
• Satellite Earth Observation (EO) is the gathering of information about the
physical, chemical, and biological systems of the planet via remote-sensing
technologies.
• Global Navigation Satellite Systems (GNSS) are a constellation of satellites

providing positioning, navigation, and timing (PNT) signals from space. GNSS
is used to track people and physical objects at any time, globally. It is also
Satellite applications industry Beyond Borders: Satellite Applications for Humanitarian Emergencies
widely used in humanitarian emergencies for geo-tagging of relevant issues

in the humanitarian context, e.g., infrastructure, disasters, damages, conflict
incidents, response activities, etc.
• Satellite Communications (SatComms) provide voice and data/internet

connectivity in regions that are not covered by terrestrial mobile networks.
• Platform and cloud services, together with technologies like machine learning,
are simplifying access to and use of satellite applications.
• New entrants and developments in SatComms and Satellite Internet of Things

(IoT) have the potential to enable new and more cost-effective connectivity to
people and things in the near future.
The use of satellite technologies and systems is referred to as an “application.” A satellite

application commonly incorporates non-satellite data and systems. Many have common
elements through shared needs, suppliers, and underlying components. Satellite
applications rely on a mix of technologies and expertise, some simple and widely
available and others advanced and specialist in nature.
Satellite applications can be a basic product or service from satellites, e.g., visual
imagery in a map to show urban infrastructure or satellite phone calls. Other
applications are the result of significant data processing and additional modelling,
e.g., asset exposure to floods within urban environments. For some applications, the
satellite data element may be small in comparison to overall activity. As introduced
earlier, suppliers are either Private Suppliers, Public Suppliers, Academia, NGOs,
Development Agencies, or a combination of these. Figure 12 describes the types of
46
services and products provided by suppliers.
Imagery Image & data Data integration
& data processing & interpretation
EC Copernicus Airbus Humanitarian

OpenStreetMap
Platform & cloud services
Google Earth Engine
Satellite & ground Position, navigation Entity & object Location, intelligence Satellite
operations & timing signals location & analysis Applications
GPS Globalstar SPOT Ushahidi
Voice & data/internet Comms & data

services equipment
Earth Satellite bandwidth Starlink

Observation & connectivity
Global Navigation
Satellite System
Inmarsat Smart modules
& devices (IoT)
SatComms
Hiber
Figure 12: Components of Satellite Applications
Satellite Earth Observation (EO)

Satellite EO is the gathering of information about the physical, chemical, and
biological systems of the planet via remote-sensing technologies. The components
of EO applications include activities for the collection, delivery, and processing of
satellite imagery and data. Data is interpreted, summarised, and presented in a way
that is accessible and useful to the intended audience using both Information and
Communications Technology (ICT) and offline technologies.
Imagery and data
For EO-based applications, basic imagery and its supply are core components. “Basic
imagery” is satellite imagery that has been acquired, downloaded, and made available
with minimal processing; this includes “raw” imagery straight from the satellite and
imagery that has been processed in several common ways, such as refining to give
a more precise location, transforming for measurement accuracy, and correcting for
various data artefacts that depend on the type of instrument or camera. These are
typically referred to as level one and level two imagery products.
47
“Analysis Ready Data” (ARD), an important development for comparative analysis
between imagery from different satellites, involves a higher level of processing. Whilst
not required for all applications, efforts like ARD decrease barriers to using multiple
sources of data and simplify the more rigorous analysis and scientific approaches
required for some applications.
An important attribute of imagery is its “resolution”: the smallest thing that can
typically be seen (resolved) on the ground. More correctly termed Ground Sample
Distance (GSD), this defines how much detail is available in imagery. Very High
Resolution (VHR) EO data is needed for many humanitarian use cases, particularly
when the emphasis shifts from broad environmental monitoring (e.g., forest change)
to observing features related to individuals and buildings (e.g., mapping informal
settlements). The resolution grades are listed below:
• Very High Resolution (VHR) (<1m)
• High Resolution (HR) (1m–4m)
• Medium Resolution (MR) (5m–25m)
• Low Resolution (LR) (25m–60m)
• Very Low Resolution (VLR) (>60m)
VHR data is more expensive than lower resolution data, as it is sourced from Private
Suppliers at commercial costs. Moreover, VHR data is also computationally more
expensive. Lower resolution imagery, such as 10 metres to 200 metres, is available from
Public Suppliers at low or no cost.
Coverage is a broad term that describes the amount of imagery and data collected
over an area or region with a given frequency. In general, Public Suppliers, such as the
European Commission’s Copernicus, have revisit times of five to eight days with global
coverage. Private Suppliers offer higher revisit times for smaller areas as standard.
For example, Planet provides new imagery at 0.5m resolution up to twelve times daily
from its SkySat constellation.48 Timely availability of data following a humanitarian
emergency (particularly in the Response Phase) is critical. Conversely, it is also
important to have access to fresh data for the Preparedness Phase, for example, in order
to quantify hazard risk scenarios. Imagery is often made available to humanitarian
organisations after an event but not preceding it.
The core delivery modes for satellite imagery are direct reception, tasked and repeat
imagery, archive imagery, and platform and cloud services, as detailed below.
Direct reception – In direct reception, a customer or partner receives imagery in near

real-time within their ground station footprint or by downloading pre-acquired stored
imagery. Direct reception within the field of view of a satellite overpass is the quickest
way to receive satellite data with latency of seconds and minutes. Direct reception
requires a ground station facility with capital outlay and ongoing maintenance for large
capital items, such as a satellite dish and reception equipment. Commercial suppliers
will typically supply licensed ground station equipment and software to facilitate
reception, simplifying implementation but at additional cost.
Tasked and repeat imagery – Tasked imagery is the requesting of single or multiple
satellite images in the future. Procurers of these services typically specify a geographical
area (Area of Interest (AOI)), timeframe, and frequency of observations for imagery.
Cloud-free or low-cloud imagery may only be possible by preparing a mosaic of a
number of images acquired over a period of time. Imagery is often supplied via online
systems from satellite operators, distributors, and resellers through web and API access.
Satellites have capacity constraints; therefore, imagery needed for a given time and
location may not always be available. For some applications and imagery types, a prior
48
arrangement to task imagery is necessary to ensure that required imagery is collected by
satellites and made available.
48 Martin Van Ryswyck, ‘Planet Announces 50 Cm SkySat Imagery, Tasking Dashboard and Up To 12× Revisit,’ Planet, June 9, 2020,
www.planet.com/pulse/tasking-dashboard-50cm-12x-revisit-announcement/
Whilst tasking is still common, imagery from some providers is not tasked and provides
regular and routine “repeat” imaging of global regions according to a defined schedule.
This approach is common for large public sector satellites (e.g., USGS Landsat series
and EC Copernicus Sentinels) and increasingly for large constellations (e.g., Planet).
Archive imagery – Many Private Satellite Operators and Public Satellite Operators
make their recent and historical imagery available via online or offline archives. Some
suppliers opt to make imagery available as soon as it is collected, downloaded, and
processed, thus reducing or removing the need for tasked imagery. Public archives of
global satellite imagery are available, including ISRO49 (India), USGS50 and NASA 51
(USA), EC52 (Europe), JAXA 53 (Japan), and INPE54 (Brazil). Regular global imagery
at resolutions generally between 10m and 200m can be acquired from these archives.
The archives typically require registration and, whilst data access is generally free and
open, terms and conditions can apply. APIs for direct access via software and apps are
also commonly provided.
Private Satellite Operators provide online and API access to their archives for imagery
that generally has much higher resolution than that of public archives. Examples
include Maxar’s SecureWatch,55 Airbus’ OneAtlas,56 and Planet’s Platform.57 In many
cases tools for viewing and using the imagery are provided; they can be easy to use
but potentially limited in comparison to the more comprehensive cloud services or
platforms described below.
For some archives, users can experience delays between the collection of imagery
and its release due to data policy or commercial reasons that differentiate the service
from tasked imagery. In many cases, priority access is available in advance or by prior
arrangement for humanitarian events. Some commercial satellite operators offer free
data access for disasters and emergencies (e.g., Planet’s Disaster Data 58 and Maxar’s
Open Data 59 Program), and the Disaster Charter60 provides rapid access to nationally
mandated disaster management authorities upon request.
International actors within the EO community (CEOS and GEO) are encouraging the
use of open data cubes to host easily accessible datasets in a standard format ready for
use. Several projects are at various stages within several countries.61
Image and data processing
Satellite imagery commonly requires processing and analysis in order to derive useful
information from it. For small numbers of images, this can often be performed on
workstations with freely available software. However, satellite imagery data tends to
require significant storage and computing power to routinely process and analyse large
datasets.
49 Indian Space Organisation National Remote Sensing Centre, ‘Bhuvan: Indian Geo-Platform of ISRO,’ https://bhuvan-app1.nrsc.gov.in/bhuvan2d/
50 US Geological Survey, ‘EarthExplorer,’ https://earthexplorer.usgs.gov
51 NASA Earthdata, ‘Earthdata Search,’ https://search.earthdata.nasa.gov/
52 ESA, ‘Copernicus Open Access Hub,’ https://scihub.copernicus.eu/dhus/#/home
53 Japan Aerospace Exploration Agency Earth Observation Center, ‘ALOS Global Digital Service Model,’
www.eorc.jaxa.jp/ALOS/aw3d30/l_map_v2003.htm
54 Instituto Nacional de Pesquisas Espaciais Earth Observation General Coordination, ‘Image Generation Division Catalogue,’
49 www.dgi.inpe.br/catalogo/
55 Maxar, ‘SecureWatch,’ www.maxar.com/products/securewatch
56 Airbus, ‘OneAtlas Data,’ www.intelligence-airbusds.com/imagery/oneatlas/data/
57 Planet, ‘Planet Platform,’ www.planet.com/products/platform/
58 Planet, ‘Disaster Data,’ www.planet.com/disasterdata/
59 Maxar, ‘Open Data Program,’ www.maxar.com/open-data
60 The International Charter Space and Major Disasters, https://disasterscharter.org/
61 Committee on Earth Observation Satellites, ‘CEOS Open Data Cube,’ www.opendatacube.org/ceos
Satellite imagery can be processed to derive physical attributes of geography within an
observed area, such as elevation,62 surface,63 and soil moisture.64 Dynamic natural and
man-made phenomena can be routinely mapped, such as fires,65 floods,66 ocean waves,67
and weather warnings.68 Imagery can be processed for features and objects extracted
so that entities such as building footprints69 and roads can be mapped and counted.70
Many commercial services now offer detailed and dynamic services, such as vehicle
and facility 71 detection.72 For supply-chain management and cargo tracking, Automatic
Identification System (AIS)73 and IoT74 services exist.
Broad indices can be derived from the values of pixels, such as for vegetation,75 water,76
and ocean colour.77 Pixel content within imagery can be classified, such as for water, a
type of vegetation, or artificial surfaces, to create maps of land cover,78 land use,79 and
urban areas.
Change detection, a common processing task, takes sets of collocated imagery and
detects changes over time. This can be used to detect new settlements, settlement
growth, or settlement removal, for example. Changes in the condition of infrastructure
can be determined for access routes.
There are a variety of software tools and techniques in the public domain to assist in
data processing.80 Machine learning (ML) is proving an increasingly important area for
satellite applications81 due to its ability to process large imagery datasets more rapidly
with a variety of learning algorithms in combination with changes in the affordability
of highly parallel processing technology. Several initiatives exist to provide access to
necessary tools and training data, notably Radiant Earth Foundation.82 Most existing
suppliers are adopting ML techniques, and new private sector companies are emerging
to specifically address opportunities, e.g., Impact Observatory.83
Data integration and interpretation
Satellite data is rarely the sole source of information required to address a humanitarian
emergency. Therefore, consideration must be given to existing sources of national
and local data and expertise, including scientific and research institutions, statistical
authorities, and domain expertise from the public and private sectors. For example, crisis
mapping organisations like the Humanitarian OpenStreetMap Team and MapAction
62 NASA Jet Propulsion Laboratory, ‘Shuttle Radar Topography Mission,’ updated 9 October 2021, www2.jpl.nasa.gov/srtm/
63 JAXA EORC, ‘ALOS Global Digital Surface Model,’ www.eorc.jaxa.jp/ALOS/en/aw3d30/
64 ESA, ‘SMOS Data—Earth Online,’ https://earth.esa.int/eogateway/missions/smos/data
65 NASA Earthdata, ‘VIIRS I-Band 375 m Active Fire Data,’ updated 7 July 2021, https://earthdata.nasa.gov/earth-observation-data/near-real-time/
firms/viirs-i-band-active-fire-data
66 Institute of Arctic and Alpine Research University of Colorado, ‘The Flood Observatory,’ http://floodobservatory.colorado.edu
67 Copernicus Marine Services, ‘Global Ocean Waves Analysis and Forecast,’ https://resources.marine.copernicus.eu/?option=com_
csw&view=details&product_id=GLOBAL_ANALYSIS_FORECAST_WAV_001_027
68 Indian Meteorological Department, ‘Warnings,’ https://mausam.imd.gov.in/imd_latest/contents/subdivisionwise-warning.php
69 European Commission, ‘Global Human Settlement Layer,’ https://ghsl.jrc.ec.europa.eu
70 Geo-Referenced Infrastructure and Demographic Data for Development, ‘GRID3 Data Hub,’ data.grid3.org
71 Aventior, ‘Insights from Aerial and Satellite Images,’ aventior.com/aerial-satellite-image-analysis/
72 4 Earth Intelligence, ‘Object and Vehicle Detection,’ www.4earthintelligence.com/products/object-and-vehicle-detection/
73 exactEarth, ‘exactAIS,’ www.exactearth.com/product-exactais
74 Hiber, ‘IoT Solutions,’ https://hiber.global
75 USGS, ‘NDVI, the Foundation for Remote Sensing Phenology,’ www.usgs.gov/core-science-systems/eros/phenology/science/ndvi-foundation-
remote-sensing-phenology?qt-science_center_objects=0#qt-science_center_objects
76 European Commission, ‘Global Surface Water Explorer,’ https://global-surface-water.appspot.com
77 NASA Goddard Space Flight Centre, ‘Ocean Color Web,’ https://oceancolor.gsfc.nasa.gov
50
78 ESA Climate Change Initiative, ‘Landcover,’ www.esa-landcover-cci.org/?q=node/164
79 OpenStreetMap, ‘Landuse Landcover,’ https://osmlanduse.org
80 ESA, ‘Sentinel Application Platform,’ http://step.esa.int/main/toolboxes/snap/; QGIS, ‘A Free and Open Source Geographic Information System,’
https://www.qgis.org/en/site/; OSGeo, ‘The Open Source Geospatial Foundation,’ www.osgeo.org
81 David J. Lary et al., ‘Machine Learning Applications for Earth Observation,’ in Earth Observation Science and Open Innovation, ed. Pierre-Phillippe
Mathieu and Christoph Aubrecht, 165–218 (Cham: Springer Open, 2018), https://doi.org/10.1007/978-3-319-65633-5_8.
82 Radiant Earth Foundation, ‘Who We Are,’ www.radiant.earth/about/
83 Impact Observatory, ‘About,’ www.impactobservatory.com/about
rely on international and local volunteers to ensure that the tracing of geographic
features in satellite imagery is correct and acceptable to affected communities.
Through the integration and interpretation of multiple data sources, a better

understanding of the risk of events emerges, for example, crop yield and productivity
forecasting for food insecurity,84 early warning systems for flood risk,85 infrastructure
mapping and exposure for multi-hazards,86 emergency response,87 and infrastructure
reconstruction.88 Given the complex nature of risk, satellite applications are increasingly
understanding and finding ways to address multiple hazards and risks.89
Platform and cloud services
Cloud services are a prominent model in the supply of satellite data by either cloud
suppliers or satellite operators. Cloud services with Infrastructure as a Service (IaaS)
make large imagery archives available within a cloud environment (storage) and provide
the IT infrastructure to run and host applications (computing). They also provide
a wide variety of supporting software tools, such as popular software development
libraries and environments, processing tools and algorithms, ML tools, and databases
and hardware (e.g., AI accelerators and sensors). Cloud services can provide quick
access to infrastructure that can be complex to independently develop and maintain
but still requires skills and expertise. Many cloud services make satellite imagery
available alongside tool sets as a standard service; examples include AWS,90 Planetary
Computer,91 and IBM Agriculture.92 What sets this model apart from “archive
imagery” is the range of additional tools, data, and services available.
Cloud providers also provide integrated ground stations, satellite operations, and data
handling services, including AWS Ground Station93 and Azure Orbital.94
Platforms provide a more specific and ready-to-use set of online tools and data that
often run on top of cloud services. Examples include Descartes Labs95 and Sentinel
Hub.96 Platforms can bring together mapping and geospatial analysis tools with online
imagery access. These include services from GIS suppliers and online mapping/virtual
globe providers, such as commercial suppliers ESRI,97 MapBox,98 Google,99 and
OpenStreetMap (OSM).100
Some platform options exist to specifically benefit humanitarian activity. OSM’s

Humanitarian OpenStreetMap Team (HOT) is a volunteer-based “international team
dedicated to humanitarian action and community development through open mapping” that
uses OSM to undertake projects and response actions across humanitarian impact
areas.101 HOT also trains and activates crisis mappers for major emergencies, like Super
Typhoon Haiyan, the Nepal earthquake, the Ebola epidemic in West Africa, and other
84 United States Agency for International Development, ‘Famine Early Warning System,’ https://fews.net
85 European Commission’s Copernicus Emergency Management Service, ‘Global Flood Awareness System,’ www.globalfloods.eu
86 British Geological Survey, ‘METEOR,’ www.bgs.ac.uk/geology-projects/geodesy/meteor/
87 Copernicus, ‘Copernicus Emergency Management Service,’ https://emergency.copernicus.eu
88 Committee on Earth Observation Satellites, ‘Recovery Observatory,’ https://ceos.org/ourwork/workinggroups/disasters/recovery-observatory/
89 Mostapha Mohammad Harb and Fabio Dell’Acqua, ‘Remote Sensing in Multirisk Assessment: Improving Disaster Preparedness,’ IEEE Geoscience
and Remote Sensing Magazine 5, no. 1 (March 2017): 53–65 https://doi.org/10.1109/MGRS.2016.2625100
90 Amazon Web Services, ‘Registry of Open Data on AWS,’ egistry.opendata.aws/tag/satellite-imagery/
91 Microsoft, ‘Planetary Computer,’ planetarycomputer.microsoft.com
92 IBM, ‘IBM Agriculture,’ www.ibm.com/products/agriculture
93 Amazon Web Services, ‘AWS Ground Station,’ https://aws.amazon.com/ground-station/
51 94 Microsoft, ‘Azure Orbital,’ https://azure.microsoft.com/en-us/services/orbital/
95 Descartes Labs, ‘Products,’ www.descarteslabs.com/products/
96 Sinergise, ‘Sentinel Hub,’ www.sentinel-hub.com
97 Esri, ‘ArcGIS Image for ArcGIS Online,’ www.esri.com/en-us/arcgis/products/arcgis-online-imagery/overview
98 Mapbox, ‘Mapbox Satellite,’ www.mapbox.com/maps/satellite/
99 Google, ‘Google Earth Engine,’ https://earthengine.google.com
100 A global crowdsourcing and mapping project similar to open knowledge movements like Wikipedia. See www.openstreetmap.org.
101 Humanitarian OpenStreetMap Team, ‘Our Work,’ www.hotosm.org/our-work
crises that get major international attention. The UN’s Global Pulse initiative also uses
big data and ML for crisis insights.102
Google’s Earth Engine “is free for research, education, and nonprofit use” and the
Amazon Sustainability Data Initiative (ASDI) “seeks to accelerate sustainability research
and innovation by minimizing the cost and time required to acquire and analyze large
sustainability datasets.” 103
Some platforms can be termed “specialised.” A specialised platform is a collection

of data and software tools that serves a particular user community with interests in
a particular domain, sector, or field, such as the Geohazards Thematic Exploitation
Platform (TEP).104
Global Navigation Satellite System (GNSS)
Position, Navigation, and Timing (PNT)
PNT signals are broadcast from a GNSS from space. These signals are augmented
to improve accuracy by ground stations for applications such as safety-critical aircraft
systems.105 Many operational GNSS satellite constellations now exist and are operated
by national governments.106 These include GPS (USA), BeiDou (China), Galileo (EU),
GLONASS (Russia), NavIC (India), and QZSS (Japan).
Whilst GNSS signals can be received globally with relative ease, augmentation
systems using ground-based infrastructure are needed for specific use cases such as
civil aviation. As augmentation systems are not yet available in some regions, the full
benefits of GNSS are not yet uniformly deployed globally.107 The EU, for example,
is cooperating 108 with African countries109 in the deployment of Satellite Based
Augmentation Systems (SBAS) for aircraft which reduce position errors for safer
landing, fuel-saving route optimisation, and airport equipment cost reduction.
Entity and object location
GNSS receivers are found in a wide variety of devices; examples include mobile phones,
farming machinery, industrial assets, and navigation systems for ships and aircraft.
PNT signals are received by these receivers to establish accurate time and position for
the object. To describe “where things are,” GNSS positioning can be used together
with local reference systems (e.g., national mapping grids and street names) to record an
object’s location both globally and locally.
An older but comprehensive EC and industry project, HARMLESS,110 found existing

GNSS humanitarian aid solutions for food distribution and tracking (use case:
population and poverty mapping; humanitarian supply chain management), mapping of
102 UN Global Pulse, ‘Crisis Insights Team,’ www.unglobalpulse.org/crisis-insights-team/

103 ‘Amazon Sustainability Data Initiative,’ Amazon, https://sustainability.aboutamazon.com/environment/the-cloud/asdi
104 Geohazards TEP, ‘Background: The Geohazards Exploitation Platform and Portal,’ https://geohazards-tep.eu/#!pages/initiative
105 ESA, ‘What is EGNOS?,’ www.esa.int/Applications/Navigation/EGNOS/What_is_EGNOS
106 US National Coordination Office for Space-Based Positioning, Navigation, and Timing, ‘Other Global Navigation Satellite Systems (GNSS),’
52 www.gps.gov/systems/gnss/
107 Space in Africa, ‘Expert Views On GNSS and Satellite-Based Augmentation Systems Initiatives in Africa,’ 23 September 2019, https://africanews.
space/expert-views-on-gnss-and-satellite-based-augmentation-systems-initiatives-in-africa/
108 European Union Agency for the Space Programme, ‘International Co-operation,’ www.euspa.europa.eu/about/international-co-operation
109 European Union Agency for the Space Programme, ‘ASECNA provides Africa’s First Early SBAS Open Service Based on the European EGNOS
Technology,’ 30 October 2020, https://www.euspa.europa.eu/newsroom/news/asecna-provides-africa’s-first-early-sbas-open-service-based-
european-egnos-technology
110 European Union Agency for the Space Programme, ‘HARMLESS,’ www.euspa.europa.eu/harmless-humanitarian-aid-emergency-management-
and-law-enforcement-support-applications
land parcels (use case: natural resource mapping), and field staff location for safety (use
case: humanitarian stakeholder coordination; search and rescue missions).111
Road transportation is critical to humanitarian logistics. Intelligent transport systems

feature communication with drivers through mobile or SatComms in combination with
precise GNSS-based location. For long routes this enables real-time rerouting for weather
and hazards, tracking of personnel and cargo, alerting for safety and security, and collecting
situational intelligence from transportation teams. For example, Globalstar’s SPOT devices
are designed to track position and notify search and rescue in the event of an emergency.112
Location intelligence and analysis
The combination of frequently updated location data for multiple assets, entities, and
objects (e.g., emergency vehicles or rescue team members in search and rescue missions)
together with timing and other datasets gives rise to location intelligence and analysis.
Also, geo-tagging relevant issues in the humanitarian context (e.g., infrastructure,
disasters, damages, conflict incidents, response activities, etc.) on the ground via GNSS
technology provides detailed contextual and positioning data to humanitarian actors.
Free and commercial platforms aimed at humanitarian and development work are
broadly available to assist in the collection, analysis, and management of location data.113
oBoToolbox, developed by Harvard Humanitarian Initiative, is a popular tool

K
used extensively in humanitarian emergencies around the world, with 132 million
survey submissions by 640,000 users in 2020.114 It is based on OpenDataKit, a fully

open-source and free data platform used by a broad range of humanitarian actors
(UNOCHA, UNHCR, ICRC, etc.), e.g., for disaster management, COVID response
coordination, and many other use cases.115
Ushahidi is a humanitarian-focused, open-source, crowdsourcing platform that

supports mapping and location data.116 Examples of usage include mapping election
violence, earthquake crisis response, and Collective Platform for Community
Resilience and Social Innovation during Crises (COMRADES).
The accuracy and management of location data is an increasingly important issue

for humanitarian organisations, as highlighted in the 2015 public discourse around
USAID’s Afghanistan health facilities data.117
The Sharing the Land project at the Université Chrétienne Bilingue du Congo uses
GNSS and GIS to map neighbourhood conflicts and land claims to assist in dispute
resolution. In improving land administration, the project seeks to address land conflict
in eastern DRC with government and local community members and “to improve
land use planning by introducing a citywide strategic planning approach that takes into
consideration the informal growth of cities in peri-urban contexts.” 118
“A GPS unit (or maybe a satellite phone with built-in GPS) is a very useful extra item,
if you have one or can get access to one.”119
111 HARMLESS Consortium, Critical Analysis Report, European GNSS Supervisory Authority, 2007, www.euspa.europa.eu/simplecount_pdf/
tracker?file=virtual_library/2007-04-19_Critical_Analysis_Report_for_the_HARMLESS_Project.pdf
112 SPOT, ‘About SPOT,’ www.findmespot.com/en-us/about-spot/company-info
113 Eyad Ghattasheh, ‘Managing Syrian Refugee Camps Using ArcGIS,’ Esri, Fall 2017, www.esri.com/about/newsroom/arcuser/managing-syrian-
refugee-camps-using-arcgis/
53
114 Personal communication to Caribou Space, 2021
115 Harvard Humanitarian Initiative, ‘KoBoToolbox,’ www.kobotoolbox.org
116 Ushahidi, ‘The Ushahidi Platform,’ www.ushahidi.com/features
117 Tom Esslemont, ‘U.S. Incorrectly Mapped Afghan Health Facilities Twice—Watchdog,’ Reuters, 24 August 2015,
www.reuters.com/article/usaid-afghanistan-gps-idINKCN0QT25120150824
118 Integrated Research Institute IRI-UCB, ‘Sharing the Land,’ https://iriucbc.org/current-projects/sharing-the-land/
119 MapAction, Field Guide to Humanitarian Mapping, 2nd ed., July 2011, https://mapaction.org/wp-content/uploads/2016/12/mapaction_field_guide_
to_humanitarian_mapping_low_res.pdf
Custom satellite applications
Mobile apps for specific use cases are increasingly available from global app stores
or sideloading sources. Safecity, for example, facilitates anonymous crowdsourced
incident reporting of sexual violence which can reach epidemic proportions during mass
demonstrations and complex emergencies.120 These datasets assist stakeholder groups
and individuals in identifying high-risk areas, and additional apps facilitate alerting and
requesting assistance.121
Satellite Communications (SatComms)
Satellite bandwidth and connectivity
Commercial and national satellite operators provide global connectivity from

telecommunications satellites. These satellites communicate with ground facilities,
called “teleports,” to make additional connections to terrestrial communications
networks, such as the public internet or telephone networks. Teleports are broadly
globally distributed, although with higher concentrations in areas of high connectivity
demand and economic activity. Satellite telecoms services have traditionally been
provided by geographically focused geostationary satellites that are “fixed” over a region
(e.g., Inmarsat122 and Intelsat123), which often also provide broadcast/television services.
Voice and data/internet services
Voice and data/internet services are now commonly available from both geostationary
satellites and numerous satellites in low earth orbit (LEO) constellations, e.g., OneWeb124
and Starlink.125 Common SatComms-based services include telephony, mobile satellite
services (MSS), and internet connectivity with all its associated digital services.
Comms and data equipment
Comms and data equipment are required to use SatComms services and include very-
small-aperture terminals (VSAT) to provide reliable voice and data connections. These
can be deployed for permanent use at a single location, but most can be easily relocated.
Small data modems provide device and computer connectivity at lower speeds, and widely
available satellite phones (handsets) often include messaging and other service options.
For emergencies, many satellite operators have signed the Crisis Connectivity Charter.
Led by the WFP and the Emergency Telecommunications Cluster (ETC),126 operators
are “committing satellite equipment and capacity that will be dedicated for humanitarian
purposes during emergency responses.” 127 The International Telecommunication Union
(ITU), a UN agency, supplies satellite phones, terminals, and telecommunications
equipment upon request to its 193 Member States during a disaster response.128
120 Human Rights Watch, ‘Egypt: Epidemic of Sexual Violence,’ 3 July 2013, www.hrw.org/news/2013/07/03/egypt-epidemic-sexual-violence
121 bSafe, ‘Feature,’ www.getbsafe.com/features; Circle of 6, ‘Home,’ www.circleof6app.com
122 Inmarsat, ‘Satellites,’ www.inmarsat.com/en/about/technology/satellites.html
54
123 Intelsat, ‘Global Network,’ www.intelsat.com/global-network/
124 OneWeb, ‘Network,’ https://oneweb.net/network
125 SpaceX, ‘Starlink,’ www.starlink.com
126 UN OCHA Inter-Agency Standing Committee, ‘Emergency Telecommunications Cluster,’ www.etcluster.org
127 EMEA Satellite Operators Association, ‘Satellite Industry and UN to Enter Operational Phase of Crisis Connectivity Charter,’ 17 May 2018,
https://esoa.net/news/satellite-industry-and-united-nations-to-enter-operational-phase-of-crisis-connectivity-charter-for-support-of-global-
disaster-relief/
128 International Telecommunications Union, ‘ITU Disaster Response,’ www.itu.int/en/ITU-D/Emergency-Telecommunications/Pages/Response.aspx
Smart modules and devices (IoT)
Smart modules are installed in small devices, IT equipment, and machinery to connect
to networks of devices in the wider Internet of Things (IoT). Communication satellites
can receive transmissions from and connect to these devices; remote IoT devices can
connect to the internet directly via satellite. Connected sensors include water-level
gauges, air pollution sensors, container-borne GPS trackers, and temperature/humidity
sensors in food storage.
Custom satellite applications
A wide variety of commercial, off-the-shelf, SatComms services and tailored

humanitarian applications are available for land, maritime, and aviation use. Rural
and remote connectivity link local authorities, community leaders, aid workers, and
citizens to regional, national, and international contacts. Communications connectivity
supports education, social and political engagement, and public service delivery
(e.g., remote medical assistance and telehealth.)129 During emergencies, remote
communications are vital for risk and incident reporting and requests for support and
assistance. Camp and logistics connectivity for digital services may include staff and
asset location tracking for safety and resource optimisation and aid worker welfare, e.g.,
messaging, social, and recreational applications.
55
129 World Health Organization, Telemedicine: Opportunities and Developments in Member States, Report on the Second Global Survey on eHealth, 2009
(Geneva: WHO, 2010), https://apps.who.int/iris/bitstream/handle/10665/44497/9789241564144_eng.pdf ?sequence=1&isAllowed=y
Landscape of
humanitarian
satellite applications
Credit: ESA
Landscape of humanitarian
satellite applications
Key points
• 42% of identified humanitarian satellite applications targeted government
Landscape of humanitarian satellite applications Beyond Borders: Satellite Applications for Humanitarian Emergencies
users.
• 62% of identified satellite applications are “customised” for a specific user.
• The highest number (92) of identified applications are for “food, security,
nutrition and famine” events in Africa.
• Many of the data gaps identified by Humanitarian

Data Exchange (HDX)
in sectors such as health and education could be addressed by satellite
applications.
To gain a quantitative picture of the landscape of satellite applications for humanitarian

emergencies, the broadest known dataset to-date has been collected and analysed.
Details of these satellite applications were derived from public sources, including
reports, web pages, news media, and outreach materials. Over 500 applications
with sufficient public information to be categorised were found. The real number of
applications is likely much larger, but this research—collecting data, gaining confidence
in coding, and analysing—was bound by time and resource constraints.
The criteria for including a satellite application in the analysis are shown in Figure 13.
Additional details of the information collected are provided in Annex 3: Methodology
and limitations.
Figure 13: Satellite Application Inclusion Criteria for Analysis
Satellite technology
Domain Use case Hazard
is core part
in scope in scope in scope
of solution
57
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Figure 14: Frequency of Satellite Applications by Country (n =>500)
Geography
Satellite applications for humanitarian emergencies are delivered globally. The dataset’s
geographical concentrations are mapped in Figure 14.
Kenya (24) is the country with the highest number of applications recorded, followed by
Bangladesh (21), Myanmar (16), Vietnam (16), India (15), and China (14).130
Whilst no countries have been explicitly excluded, the dataset was assembled with a
preference for applications in developing regions. It is also recognised that the search
was conducted in English, which may omit a significant percentage of non-English-
described applications.
Some satellite applications are present in multiple countries and are tagged “multiple,”
but not included in the map in Figure 14. Of the 127 of 248 countries with no
applications recorded, some are known to be addressed by this subset of applications.
This subset contains a variety of domains and use cases, but those related to “health
emergencies” are more prominent than in the whole dataset.
The spread of applications across regions is shown in Figure 15.

58
130 Kenya hosts two regional centres (RCMRD and ICPAC) that lead many EO-based regional projects. Kenya also hosts regional offices for the UN.
2.0%
3.1%
4.8%
8.9%
34.7%
16.9%
Asia
Africa
Multiple
LAC
29.6% Europe
Oceania
Northern America
Figure 15: Proportion of Satellite Applications by Region (%) (n =>500)

The region with the most occurrences of applications is Asia, followed by Africa and
applications that address multiple regions.
The lowest number of applications found was in Northern America. Northern America
is understood to have a high use of satellite applications as one of the world’s most
developed regions; its low percentage in the dataset is considered the result of its
economic status and lack of emergency-related applications termed “humanitarian”
rather than, for example, public services. This distinction is not applied methodically in
all regions.
Event type
Different regions have varying concentrations of satellite applications for event clusters,
as shown in Figure 16. Specific events might have a higher or lower frequency than
expected, e.g., cyclones, as explained in Annex 3: Methodology and limitations.
Event clusters with the highest number of recorded applications are:
1 Food, Security, Nutrition, and Famine (92), occurring mostly in Africa.
2 Multiple Event Types (91), occurring mostly across multiple regions. These 91
applications are mostly EO (56%), but have a larger percentage of SatComms (27%)
compared to the dataset as a whole (6.6%), highlighting the general applicability of
SatComms to multiple event clusters and use cases. Within the EO applications, many
are used for more strategic/generic purposes, such as stakeholder coordination, damage
assessment, or emergency response planning. Applications that cover Multiple Event
Types are relevant for use in analysis of compound risks and in complex emergencies
(see section Compound risks and complex emergencies).
3 Flood (83), occurring mostly in Asia.

59
Food security, Nutrition and Famine
Multiple Event Types
Flood
Infectious diseases (plant, human and animal, aquaculture)
International / Non-international armed conflict
Refugees/IDPs
Precipitation-related (Drought, Hail, Snow storm)
Illegal activity
Water supply
Environmental degradation (Land, Soil, Biodiversity, inc Forestry)
Seismogenic (Earthquake, Liquifaction)
Cyclone
CBRNE (Chemicals, Biological, Radiological, Nuclear, Explosive)
Construction/ Infrastructure / Industrial failure
Wildfire
Shelter and settlement
Chemical (Gases, Heavy metals, Pollutants, Hydrocarbons)
Marine (Storm surge, Tsunami, Ice flow)
Environmental degradation from conflict
Volcanogenic (Volcanoes and geothermal)
Insect Infestation
Terrestrial (Mudflow, Avalanche)
Transportation (Air, Road, Rail, Marine)
Complex emergencies
Post conflict recovery (Explosive remnants of war (ERW))
Invasive species
Oceania
Temperature-related (Freeze, Heatwave)
Latin America
Wind-related (Storm, Tornado) and the Caribbean
Africa
Lithometeors (Dust, Smoke)
Asia
Civil unrest Multiple
0 10 20 30 40 50 60 70 80 90 100
Figure 16: Frequency of Satellite Applications by Event Cluster and Region (n =>500)
There are a variety of event clusters less represented in the dataset, which may indicate
underserved areas requiring further investigation. However, these are hazard types that
are more common with greater sections of the global population and economy affected,
e.g., floods.
Separating applications into individual event clusters for event-based analysis is

imperfect; for example, it does not account for event chaining, when one event cascades
60
into another. However, coding applications against up to three event clusters and up to
five use cases was considered a practical approach that captures the most relevant details
for each satellite application. Alternative approaches such as impact-based analysis were
not considered practical with the time and information available.
User types
Satellite applications have demand-side users; Figure 17 shows the percentage of
satellite applications that have those user types, highlighting the relative concentration
of certain types of user.
Users of satellite applications within the dataset are predominantly Government,

NGOs, Private Sector, and Development Agencies. Lower usage is seen in First
Responders and Affected Public.
Supplier types
Applications come predominantly from Private Suppliers, followed by NGOs, Public
Suppliers, and Academia. There is little evidence of direct provision by Media and
Development Agencies.
As Private Suppliers is a large and diverse group, this is further subdivided. Supply
by Platform/Solution Providers is the most common, followed by Satellite Operator/
Resellers, Hardware/Software Suppliers, Cloud Computing Providers, and Other.
Figure 17: Proportion of Satellite Application by User Type (%) (n =>500)

3.0%
3.6
4.4%
%
4.8%
1.5%
5.5%
42.2%
Government
NGOs
9.0% Private Sector
Development agency
First responder
9.9% Affected public
General public
16.3% Academia
Media
Unknown
Figure 18: Proportion of Satellite Application by Supplier Type (%) (n =>500)

2.0%
11.8%
40.0%
20.5%
Private supplier
61 NGOs
Public supplier
Academia
24.5% Media
Development agency — 0.6%
Unknown — 0.6%
Satellite Operator/Resellers supply 25.9% of applications, highlighting extensive
vertical integration as those satellite operators move downstream, not only selling
the data streams but also developing specific applications. Conversely, as 51.2% are
Platform/Solution Providers, which are downstream-only players, they need to access
the data streams from potentially competitive upstream Satellite Operator/Resellers.
Satellite applications may service more than one type of user; Figure 20 highlights
the relationships in terms of provision of satellite applications from the supply-side to
demand-side—which is different to the count of suppliers shown in Figure 18 and 19.
Figure 20 shows that satellite applications are predominantly provided by Private

Suppliers to Government (42%) and NGOs (16%). This structure causes a barrier to
adoption in that Private Suppliers have very different missions, cultures, and processes
compared to their Government and NGO customers. This barrier is detailed in section
Mission alignment. Also, the differences between Private Suppliers and Government
and NGOs create procurement challenges; this barrier is detailed in section
Procurement challenges.
Figure 19: Proportion of Satellite Applications by Private Supplier Subtype (%) (n =>500)
2.0%
3.5%
17.4%
51.2%
25.9% Platform/solution provider

Satellite operator & reseller
Hardware/software supplier
Cloud computing provider
Other
Figure 20: Proportion of Satellite Applications in Terms of Relationship between Supplier Type and User Type (Supplier Count ≥500, Relationship Count ≥850)
43% Private Supplier Government 42%
NGOs 16%
22% NGOs
Private Sector 9%
21% Public Supplier

Development Agency 9%
62
First Responder 6%
11% Academia
Affected Public 5%
2% Media
General Public 4%
1% Unknown Academia 3%
Development Media 3%
0% Agency 1%
Unknown
Domains and events
As detailed previously, for the purposes of this report, humanitarian events are
classified according to five domains. Figure 21 shows the percentage of satellite
applications per domain.
Satellite applications occur mostly within the disasters (36.3%) domain, followed by
food insecurity (18.7%) and security and conflict (18.7%). Population displacement
(6.8%) and health emergencies (8.4%) contain the least number of applications.
11.2% of satellite applications address all domains. Of those, the vast majority also
address multiple event clusters, indicating a cross-cutting capability across domains,
events, and use cases. There are specific use cases that support all domains, for example:
infrastructure mapping and exposure, early warning systems, population and poverty
mapping, humanitarian stakeholder coordination, and connectivity for public or
remote populations. Also, from a technological point of view, SatComms applications
commonly cut across all domains and event types.
This highlights the potential of individual satellite applications to address multiple

humanitarian contexts.
Use cases
The range of addressable problems for satellite applications is broad and diverse,
as shown by the variety of use cases detailed earlier. The use cases with the most
applications found are early warning systems (128), emergency response (60), and
damage assessment (60).
Figure 21: Proportion of Satellite Applications per Domain (n ≥500)
6.8% 11.2%
8.4%
18.7%
36.3%
All
Disasters
Food insecurity
18.7%
Security and conflict
Health emergencies
Population displacement
63
Figure 22: Frequency of Satellite Applications by Use Case (n =>500)
Early warning systems
128
Emergency response
60
Damage assessment
60
Crop yield and productivity forecasting
57
Hazardous event detection and severity
53
52
46
44
33
Microplanning for health interventions
29
Refugee camps
28
Evidencing conflict-related violence

25
Extraction and trafficking of illegal resources or humans
25
Public crisis awareness
25
Subsidence and erosion
23
Crop/livestock insurance
21
Infrastructure reconstruction
17
Agricultural pest and disease response
17
Disease prevalence and transmission
16
Connectivity for public or remote populations
16
Rangelands management
14
Disaster risk financing and insurance
15
Search and rescue missions
13
Access to finance and credit
13
Contaminated sites
13
Monitoring prison camps
11
Monitoring and evaluation
10
Nuclear facilities
9
Humanitarian supply chain management
8
Election polling site monitoring
5
Compliance with peace treaties
5
64 Access routes and evacuation plans
4
Contact tracing
3
Guiding drones
3
Refugee resettlement suitability analysis
1
Business model
Supply-side actors use a variety of business models to fund the development and
operation of satellite applications, as highlighted in Figure 23.
The most common business model is a customised solution where the user pays for a
customised/bespoke application from a supplier. Open access applications are available
in the public domain to use free of cost and other access barriers. In the subscription
model, a user pays an ongoing fee for access to the service, but doesn’t own it. In research,
the satellite application has been developed primarily for research or academic purposes.
Finally, stakeholders can purchase “off-the-shelf ” and own the application from a supplier
with minimal customisation required.
There is a dominance of the customised solution business model, whereby members of

the supply chain often see the “project” as the final result, without planning or aiming
for developing a long-term, sustainable offering. This naturally leads to many one-off
implementations or pilots, which can lead to unnecessary piloting and duplication. This
barrier is detailed in section Piloting and duplication.
Figure 23: Proportion of Satellite Applications by Business Model (%) (n =>500)
11.2%
13.0% 1.6%
12.8% 61.5%
Customised solution
Open access
Subscription
Research
Off-the-shelf
65
Scale
In terms of geographies and numbers of users, satellite applications have scaled on
a spectrum, as highlighted in Figure 24. Each scale corresponds to an Application
Readiness Level (ARL). The following options are used for ‘Scale’:
• R&D applications have reached a research and development or prototyping stage

of application development, but are not yet deployed within a user’s operational
environment (ARL 1–6).131
• Implemented applications are operational services with limited geographic reach and
users (ARL 7–8).
• Scaled applications have sustained, long-term use across wide geographies, multiple
countries, and with multiple users (ARL 9). Evidence of repurposing a ‘successful’ pilot
or cross-applicability across multiple domains/events/use cases is also a contributor to
selection of this option, but is often not visible in the available literature.
The scale of each satellite application is recorded according to available literature

at a specific time and therefore will not accurately account for current situation,
continuation, and failure rates. Academia is underrepresented, as no comprehensive

review of scientific literature was conducted.
The most common scale for satellite applications is implemented. A factor in the high
proportion of Implemented projects is the many programmes that provide funding to
supply-side players for projects that are usually designed to deliver to specific users in
specific countries. These initiatives focus on supporting ‘pilot-scale implementation’ but
are not intended to offer long-term financing for resulting applications. This topic is
explored further in section Financing for application development and scaling.
Figure 24: Proportion of Satellite Applications by Scale (%) (n =>500)
16.8%
26.7% 56.6%
66 Implemented
Scaled
R&D
131 NASA, ‘The Application Readiness Level Metric,’ www.nasa.gov/sites/default/files/files/ExpandedARLDefinitions4813.pdf

Data gaps versus availability
By mapping the HDX data gaps to the satellite-enabled use cases and number of
applications identified in this report, Figure 25 illustrates humanitarian data gaps
vis-à-vis supply of satellite applications.
In the top-left of Figure 25 are use cases, including humanitarian supply chain
management, which have significant data gaps and only eight satellite applications.
Supply chain management, like many other use cases, realises indirect value from
satellite imagery but necessitates other data sources like supply inventories, warehouse
locations, needs inventories, and more. Other use cases directly benefit from satellite
data (e.g., classifying damaged buildings from imagery) and do not require additional
datasets. The bottom-right shows use cases, like early warning systems or crop yield and
productivity forecasting, that have less extensive data gaps and many more applications
currently in use.
Figure 25: Frequency of Satellite Applications by HDX Data Gaps for Each Use Case (n ≥500) Most direct value from satellite data
Indirect value from satellite data
100 High demand/low supply High demand/high supply
Data gaps (%)
Search and
rescue missions Infrastructure mapping
and exposure
90
Humanitarian supply Emergency
chain management response
80 Damage
assessment
Contact tracing
Infrastructure
70 Disease prevalence reconstruction
Refugee
resettlement and transmission
suitability Refugee camps
Guiding drones Contaminated
sites Microplanning for
Subsidence
health interventions Hazardous event
and erosion
60 Monitoring detection and severity
Access routes and evaluation Extraction and trafficking of
and evacuation illegal resources or humans
plans
Nuclear facilities
50 Access to finance and credit
Disaster risk
financing and insurance
Compliance with
peace treaties
Population
40 and poverty
Connectivity for public
or remote populations
Rangelands Crop/livestock insurance

management
30
Agricultural pest Crop yield and
and disease response productivity forecasting
Election polling Humanitarian

site monitoring stakeholder coordination Early Warning Systems
(128 applications & 20% gaps)
20
67
10
Monitoring Evidencing
prison camps conflict-related violence
Low demand/low supply Low demand/high supply

0 0 10 20 30 40 50 60 70
Number of applications
Overcoming barriers
to uptake of satellite
applications in
humanitarian
emergencies
Overcoming barriers to uptake of satellite
applications in humanitarian emergencies
Key points
• Thirteen barriers to the uptake of satellite applications in humanitarian
emergencies are identified.
• Industry, governments, and the development community are addressing each

barrier to varying degrees.
• Based on the ability of the development community to help address the issue,
a subset of eight barriers are prioritised for interventions:
− User awareness and resistance
− Inadequate monitoring and evaluation

− Data availability
− Ethics and privacy
− Technical expertise and skills
− Financing for application development and scaling
− Procurement challenges
− Piloting and duplication
• Recommended development community interventions are defined for the

eight prioritised barriers.
Barriers to uptake
This section identifies the barriers to uptake of satellite applications for humanitarian
emergencies and describes each barrier in detail. It also states whether addressing each
barrier is a priority for a development community intervention. Figure 26 identifies the
main barriers to increasing the uptake of satellite applications in humanitarian emergencies.
Figure 26: Barriers to Uptake
69 User awareness & resistance Inadequate monitoring & evaluation Inadequate understanding of needs Data availability
Regulatory constraints Ethics & privacy Technical expertise & skills Citizen safety
Financing for application

IT infrastructure Mission alignment Procurement challenges
development & scaling
Piloting & duplication

User awareness and resistance
Across the demand-side, within the humanitarian community there are specific
organisations that champion the use of satellite applications, for example, UNOSAT
within the UN system. Equally, inside organisations, there are often individual
advocates who trial and test such innovations. However, there is significant variation
across the user community in the level of awareness of satellite applications, their use
cases, and their advantages over terrestrial data sources.
Cost structures can be unfamiliar and confusing (see section Procurement challenges).
Supply-side players are often unknown to members of the humanitarian community,
and it can be difficult to understand the technical capabilities within the supply-side
(i.e., the “art of the possible”) in order to appropriately articulate the user requirements
that need to be met. Thus, satellite applications might never be fully explored as an
option for the humanitarian community, or inappropriate or over-specified technical
solutions may be developed, which are then not valued by their users (see section
Inadequate understanding of needs).
Introduction of new technology, including satellite applications, carries the potential for
human and political resistance, as such technologies change—sometimes radically—the
methods for achieving a goal. For example, if national crop yield forecasts are produced
within the Ministry of Agriculture, the national government controls them. The same
forecast produced via a satellite application provided by a national or even international
Private or Public Supplier would mean a loss of control over such statistics. Many
people still fear job loss due to emerging technologies and new methods replacing
them. If they do not have experience with satellite data, these technologies can be
intimidating and involve steep learning curves in terms of technical specifications,
appropriate use cases, business models, and more.
Trust in traditional methods has developed over many decades or centuries, and a
similar trust needs to be built with any replacement satellite-based method. The issue
of trust is often compounded by the so-called Black Box Problem with ML-derived
applications, whereby “a black box is a system which can be viewed in terms of its inputs
and outputs, without any knowledge of its internal workings.” 132 There are some efforts
tackling these transparency and replicability issues in terms of sharing guidance on
metadata standards, advocating for open source licensing, adopting frameworks based
on international human rights law and ethics, and more.
There are often numerous satellite applications that serve a similar purpose, for
example, global land cover. This makes it difficult for users to filter through the options
and know which is most appropriate for their context. Whilst there is strong supply-
side and demand-side expertise, what is lacking is ‘connector/bridge’ organisations and
persons that can link the two sides.
70
132 Wikipedia, ‘Black Box,’ updated 5 October 2021, https://en.wikipedia.org/wiki/Black_box

box 1: Overcoming User Awareness and Resistance
Addressing user awareness and resistance is a priority for development

community interventions.
Sharing knowledge and raising awareness of the opportunities, use cases, existing
examples, and outcomes and impacts of satellite applications is a public good that is
appropriate for development community interventions.
Existing initiatives addressing the awareness barrier include UKSA International

Partnership Programme,133 ESA Global Development Assistance (GDA),134
UNOOSA,135 GEO,136 and Anticipation Hub.137 However, this is a long-term barrier
which will require continued, extensive effort from multiple parties to address.
Examples of potential interventions include continuing to document existing examples

and case studies, and sharing such information via an online knowledge base and
virtual and real-world knowledge events. Future interventions should seek to go beyond
individual initiatives in order to compile a more holistic landscape of case studies.
Inadequate monitoring and evaluation
Monitoring and evaluation (M&E) is an objective process of understanding how a

project was implemented, what effects it had, for whom, how, and why. Broadly, in the
satellite applications sector, there is a lack of quality M&E and a weak evidence base
of the development outcomes and impacts of using satellite applications. Traditionally,
supply-side organisations have been focused on measuring and communicating the

scientific and technical achievements of their products, for example, the accuracy of an
EO algorithm, as opposed to quantifying the benefits for the demand-side users, for
example, in terms of lives saved in a disaster. Humanitarian organisations also need to
understand if and how satellite data can improve or add value to their decision-making,
as well as the specific circumstances in which satellite data is particularly cost effective.
In many cases, the return-on-investment thesis is not well documented or understood.
As identified in Table 3, with the use case Monitoring and evaluation, satellite
applications can directly support the M&E process. This includes “Geospatial Impact
Evaluations (GIEs) that rigorously evaluate the impacts and cost-effectiveness of specific
development interventions and large investment portfolios with spatial data. GIEs methods
leverage readily available data, like satellite observations or household surveys, to establish
a reliable counterfactual for measuring impacts—at a fraction of the time and cost of a
‘traditional’ randomized controlled trial (RCT).” 138 However, such techniques and impact
evaluations are currently rare.
133 UK Space Agency, ‘International Partnership Programme,’ www.gov.uk/government/collections/international-partnership-programme

71
134 Caribou Space, Adoption and Impact of Earth Observation for the 2030 Agenda for Sustainable Development (Farnham, Surry, UK: Caribou Space, 2020)
https://eo4society.esa.int/wp-content/uploads/2020/07/Caribou-Space_ESA-EO-for-Agenda-2030-v2.pdf
135 UN Institute for Training and Research, ‘United Nations Satellite Centre UNOSAT,’ www.unitar.org/sustainable-development-goals/united-
nations-satellite-centre-UNOSAT
136 Group on Earth Observations, ‘Disaster Risk Reduction Working Group,’ https://earthobservations.org/drr_wg.php
137 Anticipation Hub, ‘Earth Observation for Anticipatory Action (EO4AA),’ www.anticipation-hub.org/exchange/working-groups/earth-observation-
for-anticipatory-action-eo4aa/
138 AidData, ‘Geospatial Impact Evaluations,’ www.aiddata.org/gie
box 2: Overcoming Inadequate Monitoring and Evaluation
Addressing inadequate monitoring and evaluation is a priority for development

There are existing initiatives upskilling the satellite applications sector on M&E
techniques. For example, UKSA IPP has embedded robust M&E throughout the
programme design and has trained, via Caribou Space, ~50 organisations on M&E.
However, IPP worked with only a subset of organisations from the global sector. This is
a long-term barrier which will require continued, extensive effort from multiple parties
to address.
Examples of potential interventions would include publishing impact evaluations and

providing access to these via an online knowledge base, and donors ensuring these
are core project deliverables. Also providing training and capacity building to supply-
side organisations on M&E techniques adapted specifically to satellite applications.
Inadequate understanding of needs
Often, there is a misalignment between the technical solution and the nature of the
problem (user requirements). Supply-side organisations often don’t fully understand
the intricacies of the humanitarian domains, and if demand-side organisations are
not available or able to communicate specific user requirements, misalignment occurs.
This commonly manifests in over-engineering, over-customisation, and/or over-
simplification of the satellite application, leading to lengthy delivery time frames, high
costs, and complex operations/maintenance.
A related issue is “over-selling” the capabilities of satellite applications, which can lead
to inaccurate and non-validated information. This can then lead to disappointment
and mistrust in user communities when results are not achieved or even relevant for a
specific stakeholder.
In the humanitarian data ecosystem, there are intermediaries, typically multinational

NGOs, between the communities in need of humanitarian assistance and those making
decisions about how those needs are met. This produces a gatekeeper effect where
the needs and priorities of Affected Communities are not always best represented.139
However, in some situations, technology has enabled Affected Communities to further
engage in more valuable ways, both as data providers and users. Communities are
becoming involved in the Preparedness and Response Phases, especially as technology
enables improved data provision, including through, for example, crowdsourcing.
However, for many who have lived through traumatic events, volunteering time to
crowdsource insights from imagery can trigger stressful reactions.140 Humanitarians
need to be more aware of the psychological effects on such communities, especially
when members are asked to support response and recovery efforts. David Garcia, a
volunteer at Humanitarian OpenStreetMap team, highlights that care should be taken
to not separate the social and technical aspects of geospatial analysis, to ensure the trust
of communities, their leaders, and local governments, whilst also being technically
astute to conduct the technical analysis.
Much more interaction, especially between Affected Communities and data producers,
72
is needed to better determine the appropriate use of data and technology.
139 Jos Berens et al., ‘The Humanitarian Data Ecosystem: The Case for Collective Responsibility,’ Stanford Center on Philanthropy and Civil Society,
2017, https://pacscenter.stanford.edu/wp-content/uploads/2017/11/humanitarian_data_ecosystem.pdf
140 Caribou Space, ‘Beyond Borders. Community Engagement Workshop 3,’ 2021
box 3: Overcoming Inadequate Understanding of Needs
Addressing inadequate understanding of needs is not a priority for

development community interventions.
This barrier is common when any form of technology is applied to humanitarian contexts,
including, for example, mobile or other ICT-based applications. The importance of
understanding user needs is embedded into the Principles for Digital Development.141
Under these principles, donor organisations are expected to place a high priority on the
supply-side organisations to ensure robust understanding of needs via grant-funded
deliverables such as Political Economy Analysis, User Requirements Documents, and
Long-Term Sustainability Plans. Understanding user needs should continue to be a
principle for any donor-funded programme.
Given the ongoing existing initiatives to address this barrier, we do not identify it as a
priority barrier for a new development community intervention.
Data availability
Large public sector investments in initiatives such as the European Copernicus

programme have made large volumes of data openly available and free to use. However,
for some humanitarian use cases, these data sources are of insufficient spatial resolution
or may not be readily available for use during a crisis. Furthermore, despite the
increasing availability of Earth imagery from private sector actors, such as Planet and
Maxar, the cost of licensing this imagery may be prohibitive for some humanitarian
organisations; the time-intensive procurement process may also be too lengthy. In
addition, accessing data in a timely fashion may be affected by other demands and
tasking requests on the available satellites, meaning that humanitarian needs may not
be met as rapidly as required.
Within the demand-side, a lack of coordination between user organisations, and even
within single organisations, can lead to a lack of awareness of who owns or has procured
satellite applications or data. Organisational silos are common for both supply-side and
demand-side organisations, equally affecting private and public organisations. These
silos are “ business divisions that operate independently and avoid sharing information.” 142
This can lead to duplicative resource-intensive procurements of EO data, inefficient
costs, and limited usage. Further, the products are often inaccessible to local
communities. This is compounded by data-sharing agreements that often do not allow
sharing raw data.
For satellite applications to have value, they need significant data from terrestrial
sources to be combined. For example, in assessment of damaged buildings, satellite data
will not highlight how many people were inside the building. For flooding applications,
population datasets need to be combined to forecast how many people are affected by
the flood.
Figure 23 shows that the most common business model is a customised solution
where the user pays for a customised/bespoke application from a supplier (61.5%). This
highlights that there may be many duplicative satellite applications being developed for
separate users but with very similar requirements.
73
141 Principles for Digital Development, ‘Principles,’ https://digitalprinciples.org/

142 Will Kenton, ‘Silo Mentality,’ Investopedia, 24 November 2020, https://www.investopedia.com/terms/s/silo-mentality.asp
box 4: Overcoming Data Availability
Addressing data availability is a priority for development community

interventions.
Part of the cause of this barrier is a lack of visibility of who is doing what and where,
leading to coordination failures and duplication of effort. This is a classic issue across
the development community and equally in satellite applications. Addressing it
requires continued, extensive effort from multiple parties.
Examples of potential interventions would include open and free information sources
of who is doing what and where on the topic of satellite applications for humanitarian
contexts. An example of such an information source is an online open knowledge
base, and virtual or real-world knowledge events (same as for User awareness and
resistance barrier).
Other potential interventions include donors funding access to high-resolution EO

data for usage by humanitarian organisations.
Regulatory constraints
Regulation and legislation can introduce significant operational and legal barriers for
the use of satellite applications. Conversely, an absence of regulation or enforcement
is potentially undesirable; for example, unregulated spectrum use can lead to radio
interference in satellite signals.143
There are five UN Committee on the Peaceful Uses of Outer Space (COPUOS)
international treaties and five sets of principles generally beneficial to humanitarian

solutions that have been ratified by many countries.144 The UN Office for Outer
Space Affairs (UNOOSA) records the existence of 29 national space agencies145 and
a collection of 27 countries’ space laws.146 Many national space laws reference the
UN treaties and principles. However, barriers exist in national regulation influenced
by multiple policy areas, including defence and security, industrial policy, economic
development, ICT and geospatial data, innovation, and international cooperation.
A recent example of national law reducing barriers is India’s February 2021 geospatial
guidelines, which removed the need for licences and approvals for Indian organisations
to create and publish geospatial data (with sensitive exceptions).147 Indian entities are
now allowed to access ground stations and augmentation services for more accurate
GNSS real-time positioning and other previously restricted public survey datasets.
The UN International Telecommunication Union (ITU) coordinates radio-frequency

spectrum, satellite orbits, and technical standards. By agreement and convention of
the 193 member states, all satellites require registration with the ITU by national
administrations (e.g., a communications regulator or space agency). To enable
deconfliction of radio spectrum usage, the ITU reviews frequency assignment notices
submitted by administrations for formal coordination procedures or recorded in the
Master International Frequency Register (MIFR).
74
143 Jonathan Amos, ‘Europe’s Smos ‘Water Mission’ Battles Interference,’ BBC News, 5 May 2010, http://news.bbc.co.uk/1/hi/sci/tech/8661228.stm
144 UN Committee on the Peaceful Use of Outer Space Legal Subcommittee, 60th Sess., Status and Application of the Five United Nations Treaties
on Outer Space, UN Doc A/AC.105/C.2/2021/CRP.10 (31 May 2021), www.unoosa.org/res/oosadoc/data/documents/2021/aac_105c_22021crp/
aac_105c_22021crp_10_0_html/AC105_C2_2021_CRP10E.pdf
145 UN Office for Outer Space Affairs, ‘World Space Agencies,’ www.unoosa.org/oosa/en/ourwork/space-agencies.html
146 UN Office for Outer Space Affairs, ‘National Space Law,’ www.unoosa.org/oosa/en/ourwork/spacelaw/nationalspacelaw/index.html
147 Government of India, ‘Guidelines for acquiring and producing Geospatial Data and Geospatial Data Services including Maps,’ 15 February 2021,
https://dst.gov.in/sites/default/files/Final%20Approved%20Guidelines%20on%20Geospatial%20Data.pdf
These processes face considerable evolving challenges, including:
• Increasing impact of radio interference
• Limited availability of ”slots” for geostationary satellites
• Inactive, exaggerated, or updated registrations unnecessarily reserving valuable

telecoms capacity148
• Increasing collision and debris risk from satellites, constellations, and mega-
constellations in LEO
• Political challenges149
There are many standards relevant to the implementation and delivery of humanitarian
solutions (e.g., OGC,150 EU INSPIRE,151 UN-GGIM152) that address interoperability
and common mapping representations, for example. Absence of such standards
represents a barrier to uptake, transferability, and integration into wider solutions.
However, standards can be complex and difficult to implement, and may also represent
a complexity and cost barrier for humanitarian organisations.
Humanitarian organisations must also be aware of export controls, which are used
by countries “to protect national security interests and promote foreign policy objectives.” 153
Governments control the export of sensitive equipment, software, and technology for
reasons of national security and foreign policy; these regulations cover SatComms,
GNSS, and EO. Suppliers assume responsibility for compliance; for every potential
customer, they must know the intended use, end user contact details, and ultimate
destination of the data.154 Supply-side providers cannot distribute export-controlled
materials to individuals or companies in “embargoed countries.” 155 Commercial and
export controls make states having access to “available analysed information concerning
the territory under its jurisdiction” difficult to implement universally across borders.
When humanitarian emergencies occur in “embargoed countries,” it is not possible
to purchase from supply-side companies based in supplier countries with such export
controls. Organisations based in those embargoed countries are also restricted from
purchasing data from these supply-side companies due to such government-specific
regulations.
Besides specific regulations, many national governments have policies in place to direct
spending to domestic businesses; as a result, the countries that spend the most on space
programmes are home to the largest suppliers. There is an intentional bias to keeping
that spending in-country, and this directly influences countries’ technical capacity to
use satellite data.
148 Peter B. de Selding, ‘Signal Interference Proposal Could Make the ICT a Watchdog with Some Teeth,’ SpaceNews, 10 October 2014,
https://spacenews.com/42147signal-interference-proposal-could-make-the-itu-a-watchdog-with-some/
149 Elizabeth Hoffman and Kristen Cordell, ‘An Obscure UN Agency Guides Digital Communications—Congress Must Endorse US Leadership',
The Hill, 23 July 2021, https://thehill.com/opinion/international/564527-an-obscure-un-agency-guides-digital-communications-congress-must
75
150 Open Geospatial Consortium, ‘About OGC,’ www.ogc.org/about
151 European Commission, ‘INSPIRE Geoportal,’ https://inspire-geoportal.ec.europa.eu/theme_selection.html?view=qsTheme
152 UN Statistics Division, ‘Committee of Experts on Global Geospatial Information Management,’ https://ggim.un.org
153 International Trade Administration, ‘US Export Control,’ www.trade.gov/us-export-controls
154 Maxar, ‘Maxar Trade Compliance Policy,’ 2021, https://maxar-marketing.s3.amazonaws.com/files/legal/201112%20Maxar%20Trade%20
Compliance%20Policy%20-%20final%20EXTERNAL.pdf
155 United States Treasury, ‘Sanctions Programs and Country Information,’ https://home.treasury.gov/policy-issues/financial-sanctions/sanctions-
programs-and-country-information
box 5: Regulatory Constraints
Addressing regulatory constraints is not a priority for development community

interventions.
Addressing the barrier of regulatory constraints at a national level is the responsibility

of national governments. However, the need for international regulation requires
intergovernmental organisations such as UNOOSA, COPUOS, and UN ITU.
These UN agencies are part of the development community, and their interventions
are required over the long term to address this barrier. However, we do not identify
this as a barrier for development community organisations outside of the UN system.
Ethics and privacy
The many advantages of satellite applications, including coverage, repeatability, thematic

detail/resolution, and speed, also carry risks in terms of ethics and privacy. All forms
of satellite technology, EO, SatComms, and GNSS carry the risk of observing and
conveying personal information, and thus carry privacy and ethical implications. This
risk is heightened when satellite data is used in combination with other data sources
that confer even more detailed information about individuals and/or households. This is
increasingly the case with machine-learning based approaches, IoT technology, and other
data-intensive methods that leverage satellite data alongside other disparate sources.
For humanitarians, anticipating potential risks in using satellite data can also be
difficult, as many of these issues are context-specific. For example, VHR EO can
produce detailed geo-referenced datasets over populations, including vulnerable
communities, as well as military bases, peacekeeping operations, and critical

infrastructure.156 Having this data is essential for humanitarian missions, but can also
be compromising for the communities and households being imaged. SatComms can
relay sensitive personal information as well as key information from actors on the
ground of a humanitarian event. GNSS provides location data from mobile devices
which details an individual’s whereabouts. With new cloud-based GNSS, device
anonymisation is not always guaranteed.157
Informed consent is also impossible when using satellite data sources. The person or groups
involved may not be aware of this invasion of privacy or, if they are aware, are unlikely to
be able to mitigate it or easily take remedial action, in real time or retrospectively.158 Rupert
Allan, former country manager of HOT in Uganda, recommends that organisations think
about who is directing and controlling the collection of data, and to what degree it is
actually “accessible and accountable to and by local communities.” 159
Machine learning is a part of many EO-based satellite applications used to process the
data into information and insight. ML, both inside and outside of satellite applications,
has a range of ethical risks, including automated decision-making, inherent bias,
and misleading synthetic data.160 These risks can diminish trust in ML-produced
results and, given that much of the world’s ML talent resides in wealthier countries,
algorithms are biased to reflect those cultures and societal needs.
156 Matt Swayne, ‘Researchers Detail Privacy-Related Legal, Ethical Challenges with Satellite Data,’ Phys.Org, 12 July 2019,
76 https://phys.org/news/2019-07-privacy-related-legal-ethical-satellite.html
157 Günter W. Hein, ‘Location Privacy Challenges and Solutions, Part 1,’ Inside GNSS, 18 September 2017,
https://insidegnss.com/location-privacy-challenges-and-solutions/
158 G. Berman, S. de La Rosa, and T. Accone, Ethical Considerations when Using Geospatial Technologies for Evidence Generation, Innocenti Discussion Paper
2018-02, (Innocenti, Florence: UNICEF Office of Research, June 2018), www.unicef-irc.org/publications/pdf/DP%202018%2002.pdf
159 Malia Politzer, ‘Why We Need to Think about Ethics when Using Satellite Data for Development,’ Devex, 25 February 2021,
www.devex.com/news/why-we-need-to-think-about-ethics-when-using-satellite-data-for-development-99148
160 Cem Dilmegani, ‘AI Ethics: Top 9 Ethical Dilemmas of AI and How to Navigate Them,’ AI Multiple, updated 5 October 2021,
https://research.aimultiple.com/ai-ethics/
box 6: Overcoming Ethics and Privacy
Addressing ethics and privacy is a priority for development community

interventions.
Addressing ethics and privacy is not the natural focus of industry actors, and national
governments have varying standards and regulations regarding ethics and privacy. The
development community, with its focus on improving the lives of the world’s citizens,
is an appropriate group to address this barrier.
Examples of potential interventions would include documenting and sharing best

practices for both supply- and demand-side actors. Harvard Humanitarian Initiative’s
Signal Code provides a human rights-based approach to information in crises; these
are ethical obligations for humanitarian actors and minimum technical standards for
the safe, ethical, and responsible conduct of humanitarian information activities before,
during, and after a crisis.161 Locus Charter is a proposed set of common international
principles to support ethical and responsible practice when using location data.162
Modelling and upscaling community involvement in data generation through

community mapping and validation exercises to ensure and make data open-source,
available in the public space, accountable to —and also updateable by—local
communities. Models like Taqadam, a for-profit company that offers a ML platform
to help organisations optimise data training cycles, train refugees to annotate satellite
imagery. Their annotators learn basics of data science, principles of ML, and quality
control processes. While this approach does not address the open sharing of data with
community stakeholders, it does involve them in key aspects of data generation and
provides valuable employment opportunities.
Technical expertise and skills
Specific skills are required by people in supply-side organisations (see Supply-side actors),
for example a software engineer in a Platform/Solution Provider or a GIS specialist in
an NGO, to produce satellite applications. This includes skills in EO, remote sensing,
GIS, software engineering, data science, ML, and co-development with domain-specific
expertise to ensure relevance and usefulness. All these skills are highly prized, and staff
turnover following training programmes is an issue for skills retention.
Skills are also required by the demand-side actors (see Demand-side actors), such as an
analyst in Government or a First Responder, to use satellite applications. This includes skills
in model development, application operation, data interpretation, combination with other
datasets, integration of ML approaches, and linkage of outputs to existing workflows.
However, there is also the need for very basic, simple-to-use tools that support
humanitarian emergencies. An example of this is Field Papers, which allows maps to be
printed and used rapidly in emergency contexts.163 Many of these tools are built as open
source by the open knowledge community.
There are many online learning resources available that support the development of
these skills, including tools, platforms, workshops, and tutorials. For example, Radiant
Earth’s ML Hub provides such resources specifically for machine learning for EO.164
77
161 Faine Greenwood, Caitlin Howarth, Danielle Escudero Poole, Nathaniel A. Raymond, and Daniel P. Scarnecchia, The Signal Code: A Human Rights
Approach to Information During Crisis, Harvard Humanitarian Initiative, January 2017, https://hhi.harvard.edu/publications/signal-code-human-
rights-approach-information-during-crisis
162 EthicalGEO, ‘Locus Charter,’ https://ethicalgeo.org/wp-content/uploads/2021/03/Locus_Charter_March21.pdf
163 Field Papers, ‘Welcome to Field Papers,’ http://fieldpapers.org
164 Radiant Earth Foundation, ‘Machine Learning for Earth Observation,’ www.radiant.earth/mlhub
box 7: Overcoming Technical Expertise and Skills
Addressing technical expertise and skills is a priority for development

There are existing initiatives addressing the technical expertise and skills barrier,
including UKSA IPP via training of 7,200 persons and ESA GDA via Activity 3
Skills Transfer.165 There are extensive online training courses, both paid and free, for
developing or using satellite applications,166 including those provided by UNITAR.167

However, this is a long-term barrier which will require continued, extensive effort
from multiple parties to address it. Development interventions to support technical
expertise and skills are a classic public good with both direct benefits to the uptake of
satellite applications and spillover benefits to other sectors from general upskilling.
Examples of potential interventions would include developing a playbook of principles

and case studies that showcase good practice in using satellite technologies for
humanitarian emergencies.
Also, there could be support to regional centres of expertise, such as the Regional
Centre For Mapping of Resources for Development (RCMRD)168 and/or IGAD
Climate Prediction and Applications Center (ICPAC)169 in Eastern Africa,
AGRHYMET170 in west Africa, and Asian Disaster Preparedness Center (ADCP)171
and/or ICIMOD172 in Asia.
Citizen safety
Powerful technologies often carry the risk of malevolent use. Satellites transmit
sometimes sensitive data to and from Earth, making them a potential target for
hackers or malicious actors. Satellite data can be doctored to spread misinformation, for
example, with erroneous positioning, deep fake imagery, or falsified communications
intercepts. In addition to intentional misinformation, there is also the potential for
well-intended analysts to incorrectly interpret an image or release information that
undermines the well-being of populations.173 As technical barriers are lowered, there
is increasing risk that SatComms can be hacked and transmitted information can
be leveraged by bad actors, affecting military command systems, launch systems,
communications, telemetry, and tracking.174 From a security perspective, the radio
frequency interference from electronic devices, radio antennas, or modems is strong
enough to drown out the relatively weak signals to GNSS receivers. That vulnerability
creates a tempting target for bad actors.175
165 Caribou Space, Adoption and Impact of Earth Observation for the 2030 Agenda for Sustainable Development, (Farnham, Surrey, United Kingdom.
Caribou Space, 2020), https://eo4society.esa.int/wp-content/uploads/2020/07/Caribou-Space_ESA-EO-for-Agenda-2030-v2.pdf
166 UK Space Agency, Space for Smarter Government Programme, ‘Training,’ https://spaceforsmartergovernment.uk/training
167 UNITAR, ‘UN Satellite Centre UNOSAT,’ https://unitar.org/event/event-pillars/united-nations-satellite-centre-unosat
168 Regional Centre for Mapping of Resources for Development, ‘About RCMRD,’ www.rcmrd.org/about-us/about-rcmrd
169 IGAD Climate Prediction and Applications Centre, ‘About Us,’ www.icpac.net/about-us/
170 UN Office for Outer Space Affairs, UN-SPIDER Knowledge Portal, ‘AGRHYMET Regional Centre,’
www.un-spider.org/agrhymet-regional-centre
171 Asian Disaster Preparedness Center, ‘About Us,’ https://www.adpc.net/igo/contents/adpcpage.asp?pid=2
78 172 International Centre for Integrated Mountain Development, ‘Who We Are,’ https://www.icimod.org/who-we-are/
173 Steven Livingston, ‘Satellite Imagery Augments Power and Responsibility of Human Rights Groups,’ Brookings Institution, 23 June 2016,
www.brookings.edu/blog/techtank/2016/06/23/satellite-imagery-augments-power-and-responsibility-of-human-rights-groups/; Stefan Voigt et al.,
‘Towards Semi-Automated Satellite Mapping for Humanitarian Situational Awareness,’ IEEE 2014 Global Humanitarian Technology Conference,
https://elib.dlr.de/95502/1/VoigtEtAl_IEEE_GHTC_06970315.pdf
174 Airbus, ‘Protecting Everyday Life—How Airbus Protects Satellite Systems Against Attacks,’ https://airbus-cyber-security.com/news/protecting-
everyday-life-how-airbus-protects-satellites-against-attacks/
175 Ingo Baumann, ‘GNSS Cybersecurity Threats: An International Law Perspective, Inside GNSS, 3 June 2019, https://insidegnss.com/gnss-
cybersecurity-threats-an-international-law-perspective/
Malevolent actors might use satellite applications directly against either other countries
or their own citizens, for example, the use of Satellite Sentinel Project’s imagery analysis
to guide combatants’ next attack on civilians.176 The Project’s goal was to acquire and
analyse imagery that would deter further aggression prior to the Sudan referendum. By
broadcasting perceived indicators of impending aggression, SSP inserted itself directly
into events and inadvertently provided valuable intelligence to bad actors, which resulted
in the kidnapping and ransoming of Chinese construction workers.177
box 8: Overcoming Citizen Safety
Addressing citizen safety is not a priority for development community

interventions.
Similarly to regulatory constraints, addressing this barrier is the remit of national

governments, their militaries, and the UN system. We do not identify this as a barrier
for development community organisations outside of the UN system.
IT infrastructure
Infrastructure is required to use and/or produce satellite applications. This includes

access to local or cloud-based processing and storage platforms, and internet access to
EO data and the outputs of executed algorithms.
Cloud platforms, such as Google Cloud Platform (GCP), Amazon Web Services
(AWS), and Microsoft Azure, address these challenges by storing and processing large
EO datasets in the cloud rather than on local IT infrastructure. They are significantly
more cost effective and have enabled large processing that was previously not possible
without sophisticated computing clusters. However, these platforms still require
subscriptions and robust internet connections to retrieve results. Google Earth Engine
(GEE) is a cloud platform that uses GCP and provides free access for users to conduct
analyses that would otherwise require enormous resources to access (download),
store, and analyse. This data and GEE’s cloud computing capabilities are available
to scientists, researchers, and developers, lowering the cost and time of obtaining,
pre-processing, and developing products from satellite imagery and building use-case
specific applications that ease data analysis and visualisation.
“Vendor lock-in” can be an issue, whereby supply-side organisations build their initial EO
applications on a specific cloud platform and then later face issues with switching to other
providers; for example, workflows built on AWS might not necessarily run on GCP.
Whilst cloud platforms reduce storage and processing requirements, they still require a
robust internet connection to access outputs. Internet access in developing regions is a
well-documented issue, so is not a major focus of this report. Barriers to internet access
include lack of mobile coverage, poor access to smartphones or desktop computers,
government taxes raising prices on bandwidth, and many more.178
79
176 Steven Livingston, ‘Satellite Imagery Augments Power and Responsibility of Human Rights Groups,’ Brookings Institution, 23 June 2016,
www.brookings.edu/blog/techtank/2016/06/23/satellite-imagery-augments-power-and-responsibility-of-human-rights-groups/
177 Michael Blanding, ‘Inside Harvard’s spy lab: How a team of Harvard geeks is using a satellite—plus a little help from George Clooney—to rewrite
the rules of humanitarianism,’ New York Times, 29 April 2012, www.bostonglobe.com/magazine/2012/04/28/inside-george-clooney-harvard-spy-lab/
RB6fK8MUYkBn3RvWFZpPqO/story.html
178 Alliance for Affordable Internet, ‘A4AI Policy and Regulatory Good Practices,’ https://a4ai.org/good-practices/; GSMA Connected Society,
Accelerating Mobile Internet: Policy Considerations to Bridge the Digital Divide in Low- and Middle-Income Countries, May 2021,
www.gsma.com/mobilefordevelopment/wp-content/uploads/2021/05/Accelerating-Mobile-Internet-Adoption-Policy-Considerations.pdf
Many countries have invested in National Spatial Data Infrastructures to efficiently manage
geographic data, metadata, users, and tools that are interactively connected.179 However,
the majority of these are in developed countries or multi-governmental institutions, such as
the European INSPIRE Initiative, UN Spatial Data Infrastructure (UNSDI), or World
Meteorological Organization (WMO) Regional Climate Centres (RCCs).
However, cloud platforms now significantly reduce the need for these at a national
level for each individual country. Cloud platforms provide three basic services: 1)
Infrastructure as a Service (IaaS) that provides flexibility for consumer-created

software; 2) Platform as a Service (PaaS) that enables the operation of consumer-
created software with a convenient operation complexity following resource efficient
application architectures; and 3) Software as a Service (SaaS) that hides operation
complexity.180 Key characteristics of these platforms include: broad network access,
resource pooling, rapid elasticity, and capability to measure services to control and
optimise resources which are critical when processing and making available big datasets
such as satellite EO, thereby reducing the cost of physical hardware.181
box 9: Overcoming IT Infrastructure
Addressing IT infrastructure is not a priority for development community

interventions.
As detailed above, the IT infrastructure barrier is being addressed by multiple

industry efforts, for example, cloud platforms. Also, the sub-barrier of internet
bandwidth, whilst a major issue, is being addressed by industry, governments, and
the broader digital development community.
Financing for applications development and scaling
Individual satellite applications, and the supply-side organisations that produce them,
require financing. When these applications—or the core purpose of an organisation—
are considered to be of value to society (i.e., a public good), there may be a solid
rationale to support the provision of public, taxpayer-generated funding via grants.
Grants tend to be provided by governments and development agencies for a very
specific “public good” purpose. This purpose might include academic research, or work
to support sustainable development or humanitarian efforts. Typically, funding will be
required for short-term projects lasting one to three years, with reporting obligations
and M&E activities attached to encourage full accountability of the grant recipients.
For many NGOs, this project-based funding is an important contribution towards their
operating costs, with core “unallocated” funding much harder to attract.
Private companies seeking these grants may go from grant to grant, seeking funds
for specific projects, thus neglecting longer-term commercialisation prospects or
scalability. As a result, organisations reliant on publicly funded grants may have less
scope to innovate and to adopt a longer-term perspective, to the detriment of building
organisational capacity and expertise. Project funding may also be limited in scope,
with insufficient funds for needs assessments of on-the-ground organisations or to cover
salaries of highly technical staff with in-demand skills.
80
179 Wikipedia, ‘Spatial data infrastructure,’ updated 4 March 2022, https://en.wikipedia.org/wiki/Spatial_data_infrastructure

180 Nina Kratzke, ‘A Brief History of Cloud Application Architectures,’ Applied Sciences 8, no. 8 (2018): article 1368,
http://dx.doi.org/10.3390/app8081368
181 Catherine Nakalembe et al., ‘A review of satellite-based global agricultural monitoring systems available for Africa,’ Global Food Security 29
( June 2021): article 100543, https://doi.org/10.1016/j.gfs.2021.100543
box 10: Improving Financing for Applications Development and Scaling
Addressing financing for applications development and scaling is a priority for

The landscaping work in this report has highlighted that there are a number of
initiatives that have provided financing to support industry to develop and pilot
applications in humanitarian settings, resulting in a high number of applications at
the “implemented” phase. However, a challenge with this “project-based” funding

is that industry players may struggle to continue to operate a solution or take it into
a new context. This growth or scaling capital is currently limited; as a result, only
those applications for which organisations can present a clear and compelling business
case are likely to find direct financing to be adopted by a paying customer or attract
venture capital investment. In the humanitarian sector, such commercially viable
applications may be limited to those targeting private sector players, such as insurance
companies, for whom there is an obvious financial benefit. Such customers and
“investment cases” are still relatively few and far between in the humanitarian sector.
Addressing this “valley of death” with appropriate patient capital could be

accomplished by development agencies or other philanthropic organisations, through
the creation of new financing mechanisms, for example.
When satellite applications or supplier organisations are unable to attract (sufficient)

public funding, they might look to other forms of capital, including debt and equity
investment. Unlike grants, this capital will require repayment or an exchange of equity
and can be particularly difficult and/or expensive to attract. This type of financing is
typically even more difficult to find in developing regions where the existence of “venture”
and early-stage capital is limited. Here, the nascent satellite application sector may find it
difficult to access the capital they need to develop and then market their products.
Mission misalignment
As highlighted in Figure 20: Proportion of Satellite Applications in Terms of

Relationship between Supplier Type and User Type, Private Suppliers are the most
dominant supplier (43%) and Government the dominant user (42%).
The business model of Private Suppliers is focused on extracting the maximum value
from a product, so reselling an EO dataset multiple times is sensible. However, if the
user is a Government or NGO with multiple agencies, like the UN, they might be
able to reuse that dataset across many agencies. The private sector’s focus on profit also
leads to proprietary ownership in data processing algorithms, which again makes good
business sense, but potentially leads to concentrations of power within a few suppliers.
81
Satellite Operators have data licensing and ownership agreements with private and
national ground-receiving stations. There have been cases where receiving stations with
data rights have been slow to release imagery during humanitarian emergencies outside
the control of the satellite operator.182
Demand-side organisations, e.g., Government, Development Agency, and NGOs,

are united in advancing social benefits, albeit with different specific missions. These
differences in mission translate to wide gaps in the organisational behaviours and
processes of the suppliers and users. However, whilst these differences exist, Private
Suppliers are valuable members of the supply-side, and efforts should therefore be
placed on bridging the differences.
Even if supply-side and demand-side organisations use the same formal language,
huge differences in vocabulary, from scientific/technical (suppliers) to domain-specific
language (users), often lead to confusion. For example, complex disaster resilience
terminology may not be shared across organisations.
A similar set of differences have prevented traditional humanitarian stakeholders from

further engaging with defence and intelligence agencies.183 For many civil society
and non-governmental organisations, the need and desire for political impartiality
contradicts the inherently political nature of military operations. Beyond the political,
differing perceptions of security requirements and distinct modes of operating in-
country have been other issues affecting collaboration. Despite these challenges,
defence and intelligence agencies have sophisticated data capacity and have long applied
cutting-edge technologies that have relevant humanitarian applications.184
box 11: Overcoming Mission Misalignment

Addressing mission misalignment is not a priority for development community

interventions.
Organisations naturally and healthily have different missions and processes, so that
issue in itself is not directly addressable. However, it is possible for organisations
with differences in mission and processes to gain a better understanding of those
differences through ongoing engagement and collaboration.
Therefore, no specific development community interventions for this barrier are identified.
Procurement challenges
Procurement processes are ultimately the connecting and contractual link between the
supply-side and demand-side, and there are many challenges within them.
Being an intelligent customer is “the capability of the organisation to have a clear

understanding and knowledge of the product or service being supplied.” 185 It requires the user
to be able to discern between the array of satellite applications that the supply-side offers,
along with what meets their requirements and budget. This includes understanding cost
82
182 Personal communication to Caribou Space, 2021.

183 Nicolas De Torrenté, ‘Humanitarian NGOs Must Not Ally with the Military,’ Doctors Without Borders, 1 May 2006,
www.doctorswithoutborders.org/what-we-do/news-stories/research/humanitarian-ngos-must-not-ally-military
184 Tyler J. Knox, ’US Military Innovation In The 21st Century: The Era Of The “Spin-On,”’ (undergraduate thesis, Wharton School,
University of Pennsylvania, 2020), https://repository.upenn.edu/wharton_research_scholars/194
185 UK Health and Safety Executive, ‘Human Factors: Intelligent Customer Capability,’ www.hse.gov.uk/humanfactors/topics/customers.htm
ranges and variables, such as geographic coverage, data formatting, spatial resolution,
IT infrastructure costs like cloud computing, intended use, duration of access, ability to
commit to multi-year agreements, and licensing terms like derivative rights (the ability
to create data or analytics from satellite imagery or ability to publish).
While many across the supply-side freely share technical specifications and
demonstrations, they rarely publish pricing. To understand which applications are
fit-for-purpose but also affordable, demand-side actors must engage with each provider
and/or manage a time- and resource-intensive Request for Proposals (RFP) process.
Also, understanding the supply-side’s licensing terms requires procurement and legal
teams who are knowledgeable on these topics that might not be available in house.186
Companies often cannot offer lower prices to the public than those offered to
government customers. Thus pricing is driven by their primary customer base: defence
and intelligence agencies. This outside purchasing power leads to unaffordable satellite
data and pricing for the nonprofits and startups in the demand-side, as well as licensing
terms tailored to those defence and intelligence customers.187
However, most Private Suppliers have discounts for humanitarian organisations and host
open data repositories where humanitarians can leverage released datasets for disaster
response or other noncommercial purposes. Companies like Planet188 and Maxar offer
free satellite imagery to universities for research purposes, as well as support disaster
response with public releases of before and after imagery over affected communities.189
As highlighted above (see section Data availability), aggregation of demand would

lead to greater data sharing and cost efficiency. However, there are challenges when
customised legal agreements and licences are needed to provide access across multiple
organisations. Some operators are recognising the need for data sharing across
partners in humanitarian response and offering more flexible data licensing terms for
those users.190 Along these lines, PLACE is a non-profit data trust that creates and
manages hyperlocal mapping data for a membership community that is open to all
and serves the public interest.191 In the forestry sector, the Norwegian Government’s
International Climate and Forests Initiative (NICFI) is sponsoring universal access to
high-resolution satellite monitoring of the tropical forests.192 The NICFI has awarded
a US$44M (£33M) contract to EO specialists Airbus, Planet, and Kongsberg Satellite
Services (KSAT). Andreas Dahl-Jørgensen, managing director of NICFI, said, “the
offered licensing terms are particularly good in this bid. It allows everyone to access high-
resolution satellite data, without restrictions on use and distribution.” 193
186 Caribou Space, ‘Beyond Borders. Community Engagement Workshop 2,’ 2021
187 Joe Morrison, ‘The Commercial Satellite Imagery Business Model is Broken,’ Medium, 7 August 2020, https://joemorrison.medium.com/the-
commercial-satellite-imagery-business-model-is-broken-6f0e437ec29d
188 Planet, ‘Disaster Data,’ www.planet.com/disasterdata/
83 189 Maxar, ‘Open Data Program,’ www.maxar.com/open-data
190 Amazon Web Services, ‘Solving Problems with Open Data Imagery: Q&A with DigitalGlobe and HOT,’ AWS Public Sector Blog, 1 February 2017,
https://aws.amazon.com/blogs/publicsector/solving-problems-with-open-data-imagery-qa-with-digitalglobe-and-hot/
191 Stefaan Verhulst et al., ‘Establishing a Data Trust: From Concept to Reality,’ PLACE,
www.thisisplace.org/blog-1/introducingplace/establishing-a-data-trust
192 Tara O’Shea, ‘Universal Access to Satellite Monitoring Paves The Way To Protect The World’s Tropical Forests,’ Planet, 2 March 2021,
www.planet.com/pulse/universal-access-to-satellite-monitoring-paves-the-way-to-protect-the-worlds-tropical-forests/
193 Norway’s International Climate and Forest Initiative, ‘New Satellite Images to Allow Anyone, Anywhere, to Monitor Tropical Deforestation,’ 23
September 2020, www.nicfi.no/current/new-satellite-images-to-allow-anyone-anywhere-to-monitor-tropical-deforestation/
box 12: Overcoming Procurement Challenges
Addressing procurement challenges is a priority for development community

interventions.
Ensuring that procurement processes are simple and effective is the remit of supply-
side organisations. However, supporting demand-side organisations, e.g., Development
Agencies and NGOs, to be more intelligent customers isn’t always in the best interest of
supply-side organisations. Also, as detailed above for the forestry sector, bulk procurement
is a mechanism that could address this barrier.
Examples of potential interventions would include training for demand-side organisations

on satellite application procurement processes and pricing and mechanisms to aggregate
demand into a bulk procurement agreement across multiple demand-side organisations.
Piloting and duplication
Figure 23: Proportion of Satellite Applications by Business Model shows that the
most common business model is a customised solution where the user pays for a
customised/bespoke application from a supplier (61.5%). Also, Figure 24: Proportion of
Satellite Applications by Scale shows that most satellite applications are categorised as
implemented (56.6%), with only 26.7% categorised as scaled.
These statistics reflect a well-known barrier in the use of satellite applications for
humanitarian and broader development purposes: piloting and duplication. That is
where projects or products are implemented with a single user or a small group, often
in a single country, but do not scale beyond that initial implementation to become a
more broadly adopted application, often termed pilots. One-off, customised solutions or
pilots will struggle to satisfy global market demand.
Piloting and duplication is driven by multiple factors. First, members of the supply-side
often see the “project” as the final result, without planning or aiming for development
of a long-term sustainable offering.
Second, multiple funders, be they Government or Development Agency, often provide

project-based funding, leading to multiple projects that provide similar solutions,
(see section Financing for applications development and scaling) and their recipients
therefore move from one grant-based project to the next. This is often due to a lack
of transparency and visibility as to who is doing what and where, which leads to
duplication. This is also an issue in the research/academic arena, where there is a drive
to new/innovative research, with less funding available for scaling proven concepts.
Finally, the greater the need for local customisation of a solution, e.g., the need to
combine large volumes of local ancillary data, or for local languages, the more likely
this barrier will be an issue.
84
box 13: Overcoming Piloting and Duplication
Addressing piloting and duplication is a priority for development community

interventions.
Whilst piloting and duplication is natural, effort should be made where possible to avoid
unnecessary duplication. Multiple applications developed for the same and/or different
stakeholders can cause confusion for end users (e.g., deciding what application to use). It is
also important for donors to consider the long-term sustainability and scalability of their
investments and for financing to align well to operational needs beyond the shorter-term
testing of new ideas and innovations.
Examples of potential interventions to reduce duplication of applications would include

improved information sources of who is doing what and where on the topic of satellite
applications for humanitarian contexts. An example of such an information source is an
online open knowledge base, and virtual or real-world knowledge events (same as for User
awareness and resistance barrier). Establishing more openness among the humanitarian
community and potentially encouraging shared access to data could also address the
duplication issues (see Data availability).
Priority barriers for development community interventions

As detailed above, Figure 27 identifies those barriers which are defined as priorities for
Figure 27: Priority Barriers to be Addressed by Development Community Interventions
User awareness & resistance Inadequate monitoring & evaluation Inadequate understanding of needs Data availability
Regulatory constraints Ethics & privacy Technical expertise & skills Citizen safety
Financing for application

IT infrastructure Mission alignment Procurement challenges
development & scaling
Piloting & duplication
85
Table 4: Summary of Priority Barriers and Recommended Development Community Interventions
Use Case Recommended Development Community Interventions
User awareness • Document existing examples and case studies

& resistance • Share information via an open, online knowledge base
Inadequate monitoring • Publishing impact evaluations via an online knowledge base

& evaluation • Donors ensure impact evaluations are project deliverables

• Training to supply-side organisations on M&E techniques
Data availability • Online open knowledge base of who is doing what and where
• Donors to fund high-resolution EO data for use by humanitarian organisations
Ethics & privacy • Documenting and sharing ethics and privacy best practices
• Upscaling community involvement in data generation through community mapping and validation exercises
Technical expertise • A playbook of principles and case studies that showcase good practice
& skills • Support to regional centres of expertise
Financing for application • Appropriate patient capital from development agencies and philanthropic organisations through the
development & scaling creation of new financing mechanisms
Procurement challenges • Training for demand-side organisations on satellite application procurement processes and pricing
• Mechanisms to aggregate demand into a bulk procurement agreements across multiple demand-side
organisations
Piloting & duplication • Information of who is doing what and where via an online knowledge base
• Encouraging shared access to data
86
Conclusions
Conclusions
This report describes the current landscape of satellite applications being used in
humanitarian emergencies and provides a ‘state of play’ snapshot of a rapidly evolving
sector. Satellite applications developed by an increasingly complex web of supply-side
stakeholders can be used across a wide range of use cases throughout the life cycle of an
emergency to improve our understanding of hazards, assessment of vulnerabilities, and
deployment of capabilities.
However, despite a large number of applications that have been piloted in the
humanitarian sector, there is a limited body of evidence to offer humanitarians
guidance on where satellite technology can be used most cost-effectively and with
the best outcomes for affected populations. Furthermore, rapid innovation and
increasing piloting has led to some duplication of effort on the supply-side, and the user
community has not yet fully acquired an understanding of the potential relevance for
their work and the capabilities of the technology. Regulatory and legal frameworks have
also struggled to keep pace with new developments.
There is now an opportunity for public and private sector stakeholders to reflect on
what more could be done to increase the use of, and impact derived from, satellite
applications in humanitarian assistance. Over the coming months, the Beyond Borders
project will be developing recommendations on what actions could be taken and the
potential outcomes for such interventions.
Conclusions
88
Annexes
Annex 1
Philippines Typhoon Haiyan case study
David Garcia, a team member, has documented a specific case study regarding
Philippines Typhoon Haiyan – a summary of which is below.
Satellite applications and other geospatial technologies continue to be used in preparing

for, dealing with, and recovering from crises. A major turning point in the use of such
social and technical practices was around Super Typhoon Haiyan (Philippine name:
Yolanda) in 2013. The cyclone cut through the central part of the archipelago. The
disaster caused thousands of deaths and tremendous damage and loss to ecosystems,
livelihood, and property.
Before, during, and after the disaster, hundreds of mapping projects, exercises,
and research studies were undertaken to assist in the planning, response, and
reconstruction. Many of these projects were done in coordination with Filipino
mapmakers, emergency professionals, and internet users who dealt with both the social
and technical aspects of making humanitarian geospatial work useful for, and relevant
to, affected populations.
Annex 1 Beyond Borders: Satellite Applications for Humanitarian Emergencies
The case study provides an overview of how geospatial technologies, including those
related to satellites, EO, and remote sensing, can be useful in a major disaster like Super
Typhoon Haiyan. This includes the use of geohazard mapping and satellite systems,
crisis dashboards and government geoportals, and community, non-profit, and private
sector mapping. The case study also highlights three key issues in the production of
geographic information. These are detailed in the full case study to be published in
2022 and summarised here:
• Accuracy – The accuracy of the spatial data is important in terms of coverage, error,
damage, and other aspects of how features like buildings and roads were traced. In this
challenge, higher-resolution imagery is demanded to increase the quality of spatial data.
• Accessibility – Lack of access, complex software systems, and poor internet connections
impede the usability of the maps for the communities who directly respond to crises.
• Accountability – Continuous support must be given to open knowledge communities

who create, maintain, curate, care for, and disseminate geospatial data. The open
knowledge communities fulfil an important role of accountability to affected
populations.
90
Annex 2
East Africa case study
Stella Chelangat Mutai, a team member, has documented a specific case study
regarding East Africa – a summary of which is below.
Geospatial and satellite-based technologies have increasingly become popular in the

humanitarian community within Eastern Africa (EA). In recent decades, the increase
in humanitarian crises caused by natural disasters such as drought and famine in EA
has led to high demand for the use of satellite image mapping and for rapid situational
awareness.194 However, turning this data into relevant geospatial information products
for humanitarian actors remains a major challenge within the region, as reports are
often inaccurate, in conflict with other reports, or incomplete in terms of geographical
and temporal details.195
This case study analyses the use of satellite applications in the humanitarian context in EA.
It also seeks to understand stakeholder engagement and contributions by highlighting gaps
and determining possible entry points in the use of satellites in humanitarian decision-
making and policy-driven assessments within the identified countries. This research
focused on the following EA countries: Ethiopia, Democratic Republic of Congo, Kenya,

South Sudan, Tanzania, and Uganda. An online survey was used to understand current
activity and capacity to use satellite applications from humanitarian organisations within
the EA region. The outcome of our online survey focused on:
• Potential use cases in which satellite data uptake can be harnessed
• The level of awareness of technical specifications of data required by humanitarian

organisations
• The level of capacity and knowledge within EA to identify modalities of training that
are needed
Climate Change, Land Cover, Natural Hazards, and Food Insecurity, respectively, were
voted as the potential use case domains in which increasing satellite technology capacity
would have great impact. This was influenced by the fact that most domains affect day-
to-day livelihoods of people within the case study region, and there is a dire need for
efficient early warning systems and actionable information for decision-making.
The research also sought to understand the level of awareness of existing satellite
applications and the technical specifications of satellite data within existing
humanitarian organisations. The survey found:
• 73% of humanitarian organisations within the scope of study have a broad knowledge of
91
existing data and use cases; however, 27% have limited to no information.
194 Rhoda Margesson et al., Horn of Africa Region: The Humanitarian Crisis and International Response 2012, US Congressional Research Service,
12 January 2012, https://sgp.fas.org/crs/row/R42046.pdf; UN OCHA, Global Humanitarian Overview 2022, https://gho.unocha.org/
195 John Bryant, Digital mapping and inclusion in humanitarian response, Humanitarian Policy Group Working Paper, October 2021,
https://cdn.odi.org/media/documents/Digital_IP_Mapping_case_study_web_EDSoP6n.pdf
• 68% of the humanitarian organisations surveyed understand the technical specifications
of satellite applications required for different use cases; however, 32% have little to zero
capacity and awareness.
The respondents to the survey identified a number of thematic domains in which they
considered the use of satellite applications to be paramount. However, there is a gap
in the level of awareness of the technical specifications which relies on professionals
having the technical expertise, knowledge/awareness of satellite data, and frequent use
of satellite information during relief efforts.
To understand how to bridge this capability gap, the survey asked respondents to assess
their perceived requirements for training on satellite data knowledge within their
organisation. The survey found:
• 40% preferred to receive advanced level training
• 44% requested intermediate level training
• 16% requested introductory level training
This study recommends the institutionalisation of capacity building to support

proficiency in the development of satellite applications and awareness of new
applications within humanitarian organisations. Additionally, extending the
availability, access, and distribution of data and information portals would reduce data
redundancy and might enhance collaboration during humanitarian aid response. The
survey suggested that there is a need for partnerships that bring technical professionals
together with domain experts who can champion, and demonstrate the use of EO in
relevant thematic areas across the humanitarian sector.
92
Annex 3
Methodology and limitations
General methodology
For this report, the team used desk-research to conduct the landscape of satellite
applications for humanitarian emergencies. Information has been gathered from sources
including academic papers, grey literature, organisation’s websites, media publications, etc.
External input and review were gathered from:
• Stakeholder Workshops. Three virtual workshops were held with over 40 participants
from 20 countries to gain their experiences and challenges with using satellite
technologies in humanitarian situations.
• Advisory Group. Two Advisory Group meetings were held to gain feedback on scope,
findings and conclusions. See the acknowledgments for the members of this group.
Methodology for section landscape of humanitarian satellite applications

A Google Sheet database was created to capture consistent attributes for each
application. The team found and categorised over 500 satellite applications. Caribou
Space intends to make this information available publicly in 2022.
The inclusion criteria were:
• The satellite application applies to the domains, use cases and hazards, in scope
• The satellite application includes satellite technology as a core component of the

solution, not a peripheral part
The following attributes were used:
• Organisation – the primary provider of the application
• Product or Project Name – the commonly used name of the application
• Source URL – a website reference
• Country – the International Organization for Standardization (ISO) short English

name with the addition of multiple. A maximum of five individual countries are
recorded for each application.
• Region – the UN Statistic division (UNSD) geoscheme and Unknown196

93
• User Organisation Type – as Government, First Responder, Development Agency, Media,
Private Sector, Academia, NGOs, Affected Public, General Public, and Unknown
196 Wikipedia, ‘United Nations Geoscheme,’ updated 4 October 2021, https://en.wikipedia.org/wiki/United_Nations_geoscheme

− It was not possible to meaningfully disaggregate the classes further (e.g., Government
subcategories) given the information available for applications.
− A maximum of three user organisations are recorded.
− See details in section Supply-side actors.
• Supplier Organisation Type – as Private Supplier, Public Supplier, Academia, NGOs,

Media, Development Agency, or Unknown
− Private Supplier is sub-categorised into Satellite Operator & Reseller, Cloud Computing
Provider, Platform/Solution Provider, Hardware/Software Supplier, and Other.
− Public Supplier is sub-categorised into Satellite Operator and Analytics Provider.
− See detail in section Demand-side actors.
• Domain – either Disasters, Security and Conflict, Food insecurity, Population displacement,
Health emergencies, and All (see section Domains of humanitarian emergencies)
• Event Cluster – up to three event clusters
• Use Case – up to five use cases (see section Use of satellite applications across the event
life cycle)
• Satellite Technology Used – EO, GNSS, SatComms, SatComms – IoT, or Multiple
• Business Model
− Off-the-shelf, meaning a customer can purchase and own the satellite application from
a supplier with minimal customisation required.
− Subscription, meaning a customer pays an ongoing fee for access to the service, but
doesn’t own it.
− Customised solution, meaning a customer pays for a customised/bespoke satellite

application from a supplier.
− Open Access, meaning the application is available free of cost or other access barriers.
− Research, meaning the application is completed as research/academic study only.
− Unknown.
• Scale
− Scaled, meaning sustained, long-term use of application across wide geographies or

multiple countries with multiple users at Application Readiness Level (ARL) 9.
− Implemented, meaning operational use of application within limited geographic reach

and limited number of users at ARL 7–8.
− R&D, meaning R&D and prototyping stage of application development, not yet
deployed within a user’s operational environment (ARL 1–6).
− Unknown
This data collection methodology has the following limitations and challenges:
• The dataset is a non-probability sample based on the judgement sampling of

contributors. As a first iteration, it is not exhaustive or saturated. Where possible,
94
generalisations about the supply of satellite applications for humanitarian emergencies
are shown, but not statistical conclusions. The dataset will inevitably include biases;
these are identified where possible.
• Desk research searching was conducted with English keywords, which may limit
visibility to applications in specific regions in Latin America (Spanish dominant) and
West Africa (French dominant).
• Sources didn’t always have enough detail to define all attributes.
• It is a ‘snapshot’ of solutions developed in the recent past, in response to humanitarian

emergencies over that period.
• For some attributes, use of ‘multiple’, ‘all’, and ‘unknown’ creates issues in charting and
analytics.
95
Annex 4
Glossary
AOI – Area of Interest is the geographic extent GIS – A geographic information system is a system
of analysis designed to capture, store, manipulate, analyse,
manage, and present spatial or geographic data
API – Application Programming Interface
GNSS – Global Navigation Satellite Systems are
ARL – Application Readiness Level
a constellation of satellites providing signals from
AWS – Amazon Web Services space that transmit positioning and timing data to
GNSS receivers
CEOS – Committee on Earth Observation
Satellites Hazard – Circumstance, event or entity with
potential to cause harm or loss
Copernicus – The European Union’s Earth
Observation Programme, looking at our planet HDX – Humanitarian Data Exchange
and its environment for the ultimate benefit of all
HR – High resolution EO data with a resolution
European citizens
of 1m–4m
Demand-side – Community of global
HIH – Humanitarian Innovation Hub

humanitarian data users
Humanitarian event – A humanitarian event is a
Domain – A domain is a broad grouping for
potential starting point for a humanitarian crisis
different humanitarian crises with common
and the need for humanitarian action.
characteristics using simple naming and common
terminology Humanitarian emergencies – Are events or
series of events that represent a critical threat
DRR – Disaster Risk Reduction
to the health, safety, security or wellbeing of a
EO – Earth Observation is the gathering of community or other large group of people
information about the physical, chemical, and
IaaS – Infrastructure as a Service
biological systems of the planet via remote-sensing
technologies IDP – Internally Displaced Persons
ESA – European Space Agency IPP – International Partnership Programme,

a five-year programme run by the UK Space
GEO – The Group on Earth Observations (GEO)
Agency. IPP uses the UK Space sector’s research
is an intergovernmental organisation working to
and innovation strengths to deliver a sustainable,
improve the availability, access and use of Earth
economic or societal benefit to developing
observations for the benefit of society
countries.
Geostationary orbit – A geostationary orbit is a
IOT – Internet of Things describes physical
circular geosynchronous orbit ~35,000 kilometres
objects that are embedded with sensors, processing
in altitude above Earth’s equator and following
96 ability, software, and other technologies, and that
the direction of Earth’s rotation
connect and exchange data with other devices and
GCP – Google Cloud Platform systems over the Internet or other communications
networks
GEE – Google Earth Engine
GSD – Ground Sample Distance

LR – Low resolution EO data with a resolution Risk – The probability of an outcome having a
of 25–60m negative effect on people, systems or assets
LEO – Low Earth Orbit is an Earth-centred orbit Satellite application – Digital services and
close to the planet, often specified as an orbital products that serve a number of functions for
period of 128 minutes or less. society, the environment and the economy,
deploying satellite technology
M&E – Monitoring & Evaluation is an objective
process of understanding how a project was SatComms – Satellite communications is the
implemented, what effects it had, for whom, use of satellites to provide communication links
how and why between various points on Earth
ML – Machine Learning Supply-side – Providers of satellite applications for

humanitarian emergencies
MSS – Mobile satellite services provide two-way
voice and data communications to global users UK FCDO – Foreign, Commonwealth and
who are on the go or in remote locations Development Office
MR – Medium resolution EO data with a Use case – A specific problem for the demand-side
resolution of 5m–25m users that is addressed by a satellite application
ODA – Official Development Assistance is a SAR – Synthetic Aperture Radar (SAR) satellites
term defined by the Development Assistance are a form of radar using the motion of the
Committee (DAC) of the Organisation for satellite to create a larger ‘virtual’ antenna that is
Economic Co-operation and Development used to create two or three-dimensional images of
(OECD) to measure aid objects, such as landscapes
OSM – OpenStreetMap UKSA – UK Space Agency
PNT – Position, Navigation & Timing signals UNOSAT – The United Nations Institute for
broadcast from a global navigation satellite system Training and Research (UNITAR)’s Operational
(GNSS) from space Satellite Applications Programme
Resilience – The ability of a system, community VHR – Very High Resolution EO data with a
or society exposed to hazards to resist, absorb, resolution of (<1m)
accommodate, adapt to, transform and recover
VLR – Very Low Resolution EO data with a
from the effects of a hazard in a timely and
resolution of (>60m)
efficient manner, including through the
preservation and restoration of its essential basic VSAT – Very-small-aperture Terminals are
structures and functions through risk management small-sized earth stations used to transmit
and receive data, voice, and video signals over
Response – Actions taken directly before, during
a satellite communication network, excluding
or immediately after a disaster to save lives, reduce
broadcast television
health impacts, ensure public safety and meet the
basic subsistence needs of the people affected
97
© 2022 Caribou Space
Caribou Space supports organisations to bridge the space and

development worlds by working with governments, space agencies,
development agencies, and private sector space companies.
Readers are encouraged to reproduce material from Caribou Space

reports for their own publications, as long as they are not being
sold commercially. As copyright holder, Caribou Space requests due
acknowledgement and a copy of the publication.
www.caribou.space contact@caribou.space
Caribou
Space

Beyond Borders: Satellite Applications For Humanitarian Emergencies

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Beyond Borders: Satellite Applications For Humanitarian Emergencies

Uploaded by

Copyright:

Available Formats

Beyond Borders:

Satellite Applications for

The following authors wrote this report:

For further information please contact Nicki McGoh

This report was commissioned by the UK’s Humanitarian Innovation

The primary authors were supported by team members from the

• Satellite Applications Catapult: Delia Di Filippantonio, Andy

• UNITAR: Einar Bjorgo

Caribou Space acknowledges the support offered by staff from the

• Columbia University: Andrew Kruczkiewicz

• Humanitarian OpenStreetMap Team: Rebecca Firth

• UK Space Agency: Ray Fielding

• World Bank: Bernhard Metz

Caribou Space, Beyond Borders: Satellite Applications for Humanitarian Emergencies,

Published August 2022

• Official Development Assistance (ODA) fund and programme strategy: Strategic

• Research, communications, and knowledge sharing: Conducting research on market

• Programme management: Delivery of complex, multi-country, multi-million-pound

• Product strategy: Supporting strategy for the sustainability and commercialisation of

For further information on Caribou Space see

The UK Humanitarian Innovation Hub is a pilot programme, funded by the UK

This report includes the following:

• A categorisation of supply- and demand-side stakeholders of satellite technology for the

• A consolidated overview of examples where satellite technologies have been used

• A landscaping of over 500 humanitarian satellite applications.

• Identification of key barriers to the adoption of satellite applications and potential

Humanitarian emergencies share common characteristics related to the domains in

Satellite applications industry

Satellite applications utilise three types of satellite technology:

• Global Navigation Satellite Systems (GNSS) are a constellation of satellites providing

• Satellite Communications (SatComms) provide voice and data/internet connectivity in

The improving awareness and acceptance of satellite applications is leading to broader

• More timely decision-making through real-time and predictive modelling

• Decision-making supported by greater accuracy of data

• Greater confidence in the decision-making process

• Greater accountability across stakeholders

Ecosystem of satellite applications for humanitarian emergencies

There is a complex network of stakeholders involved in satellite applications for

Supply-side organisations include:

• Public suppliers: satellite operators, and analytics providers

Demand-side organisations include:

• 42% of identified humanitarian satellite applications targeted government users.

• 62% of identified satellite applications are “customised” for a specific user.

Overcoming barriers to uptake of satellite applications in humanitarian emergencies

Thirteen barriers to the uptake of satellite applications in humanitarian emergencies

• User awareness and resistance

• Inadequate monitoring and evaluation

• Ethics and privacy

• Technical expertise and skills

• Financing for application development and scaling

• Piloting and duplication

Barrier Recommended Development Community Interventions

User awareness & • Document existing examples and case studies

Inadequate monitoring • Publish impact evaluations via an online knowledge base

Satellite applications are being developed by an increasingly complex web of supply-

• UKHIH selected a consortium of organisations to author this report, including

Institute for Training and Research (UNITAR).

Humanitarian Innovation Hub

UKHIH is leveraging world-class UK science, technology, and innovation to support

By facilitating a network of UK humanitarian innovators focused on new and improved

UKHIH selected a consortium of organisations to support this task, including Caribou

Satellite Earth Observation

Satellite Communications (SatComms) is the use of satellites to provide